032006 02_RISTROPH.DOC
4/24/2006 12:27 PM
PROPORTIONALITY AS A PRINCIPLE
OF LIMITED GOVERNMENT
!
ALICE RISTROPH†
ABSTRACT
This Article examines proportionality as a constitutional limitation
on the power to punish. In the criminal context, proportionality is
often mischaracterized as a specifically penological theory—an ideal
linked to specific accounts of the purpose of punishment. In fact, a
constitutional proportionality requirement is better understood as an
external limitation on the state’s penal power that is independent of
the goals of punishment. Proportionality limitations on the penal
power arise not from the purposes of punishment, but from the fact
that punishing is not the only purpose that the state must pursue.
Other considerations, especially the protection of individual interests
in liberty and equality, restrict the pursuit of penological goals.
Principles of proportionality put the limits into any theory of
limited government, and proportionality in the sentencing context is
just one instance of these limitations on state power. This
understanding of proportionality gives reason to doubt the assertion
that determinations of proportionality are necessarily best left to
legislatures. In doctrinal contexts other than criminal sentencing,
proportionality is frequently used as a mechanism of judicial review
to prevent legislative encroachments on individual rights and other
exercises of excessive power. In the criminal sentencing context, a
constitutional proportionality requirement should serve as a limit on
penal power.
Copyright © 2005 by Alice Ristroph.
† Associate Professor, University of Utah, S.J. Quinney College of Law. A.B.,
Harvard College; J.D., Harvard Law School; Ph.D., Harvard University; L.L.M., Columbia
Law School. Thanks to all who have offered suggestions and comments, including Francis
Allen, Paul Cassell, Bryan Fair, Angela Harris, Jameel Jaffer, Brian Kalt, Erik Luna, Gerard
Lynch, Dan Markel, Dan Medwed, Melissa Murray, Susan Sturm, Jeremy Waldron, and
Franklin Zimring.
032006 02_RISTROPH.DOC
264
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
TABLE OF CONTENTS
Introduction............................................................................................. 264
I. The Limits of Penological Theory ................................................ 271
A. Utilitarian Proportionality ....................................................... 272
B. Just Deserts, Lex Talionis, and Other Retributive Ideals .... 279
C. Political Proportionality ........................................................... 284
II. Proportionality in Constitutional Doctrine ................................. 292
A. Beyond Punishment .................................................................. 292
B. Proportionality and the Pocketbook....................................... 297
C. Death, Prison, and Gross Disproportionality ........................ 300
III. Beyond Penology: Skepticism and Conduct................................ 314
Conclusion ............................................................................................... 330
INTRODUCTION
Incarceration and capital punishment involve direct exercises of
force against the human body that occur almost nowhere else in
domestic politics in a liberal state. Because liberal democratic
governments trace their legitimacy to something other than superior
physical force, and because liberal democratic governments claim to
protect the lives and liberties of their subjects, the imposition of these
sanctions may be one of the most illiberal practices of a liberal state.
For a variety of reasons, penal sanctions are often practically and
perhaps morally necessary. But punishment’s inherent tension with
liberal ideals suggests a need for principled restrictions on the scope
of the penal power.
In fact, few such restrictions apply to American sentencing
practices. Nominally, the Supreme Court considers the Eighth
Amendment’s prohibition of “cruel and unusual punishments” to
require that penalties be proportionate to offenses.1 But that
proportionality requirement has been attenuated in recent years.
Only a minimal proportionality principle seems to restrict a legislature
as it prescribes the range of sentences for a given crime. In March
2003, the Supreme Court found that lengthy prison sentences
mandated by California’s “three strikes” law did not violate the
proportionality requirement of the Eighth Amendment, even if those
1. The proportionality requirement of the Eighth Amendment was first recognized by the
United States Supreme Court in Weems v. United States, 217 U.S. 349, 377 (1910) (striking down
a criminal sentence as “cruel in its excess of imprisonment”); see also id. at 367 (“[I]t is a precept
of justice that punishment for crime should be graduated and proportioned to offense.”).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
265
sentences were imposed for seemingly minor offenses such as
shoplifting three golf clubs or the theft of approximately $150 worth
2
3
of children’s videotapes. In Blakely v. Washington and United States
4
v. Booker, the Court found some aspects of state and federal
sentencing guidelines—guidelines that originated as legislative efforts
to ensure proportionality in judicial sentencing—unconstitutional
5
under the Sixth Amendment. Although sentencing guidelines are
typically aimed more at consistency (proportionality relative to other
sentences) and less at restrictions on severity (proportionality relative
to the instant criminal offense), the concerns of the Blakely and
Booker Courts about which decisionmakers are involved in the
sentencing process are linked to concerns about the severity of
criminal sentences. In both cases, marginal increases in severity
triggered the Court’s constitutional scrutiny of the defendants’
sentences.6 Further, regardless of one’s view of the conceptual
underpinnings of the Court’s rejection of mandatory sentencing
guidelines, that rejection provides the occasion for extensive revisions
to federal and state criminal sentencing law. Rethinking sentencing in
the wake of Blakely and Booker should prompt reconsideration of the
constitutional status of proportionality.
This Article examines proportionality as a limitation of state
power in a constitutional liberal democracy. One source of confusion
2. E.g., Lockyer v. Andrade, 538 U.S. 63, 77 (2003) (upholding a sentence of two
consecutive terms of twenty-five years to life for petty theft of nine videotapes); Ewing v.
California, 538 U.S. 11, 30–31 (2003) (upholding a sentence of twenty-five years to life for felony
theft of golf clubs).
3. 124 S. Ct. 2531 (2004).
4. 125 S. Ct. 738 (2005).
5. The Court’s somewhat unusual Booker decision did not strike down the Federal
Sentencing Guidelines altogether. Id. at 738. One majority found that the Sixth Amendment
prohibited applications of the Guidelines to increase a defendant’s sentence solely on the basis
of facts found by a judge rather than a jury. Id. at 749–51. A second majority held that the
appropriate remedy was to strike down only the provision of the Sentencing Reform Act
making the Guidelines mandatory; under the newly advisory Guidelines system, the maximum
penalties for most federal offenses are much higher and judicial fact-finding does not increase
sentences above an otherwise applicable legal ceiling. Id. at 756–57. Justice Ginsburg was the
only Justice in both majorities.
6. As explained in greater detail in Part III, infra, recent Sixth Amendment sentencing
decisions promise at the minimum certain procedural protections against disproportionate
sentences. United States v. Booker, 125 S. Ct. 738 (2005); Blakely v. Washington, 124 S. Ct. 2531
(2004); Apprendi v. New Jersey, 530 U.S. 466 (2000). The same decisions may also assume a
substantive proportionality requirement, but that reading is more debatable, and the
constitutional home of a substantive proportionality requirement is more likely to be the Eighth
Amendment or the Due Process Clauses of the Fifth and Fourteenth Amendments.
032006 02_RISTROPH.DOC
266
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
about proportionality in the criminal context has been the
characterization of proportionality as an ideal linked to particular
theoretical accounts of the purpose of punishment—usually,
retributive accounts. In fact, proportionality is better understood as
an external limitation on the state’s power to incarcerate or execute
individuals, and this limitation applies whether the state is punishing
to exact retribution, to deter, to incapacitate, or (as is most often the
case) to pursue some amalgam of ill-defined and possibly conflicting
purposes.
Much turns on whether proportionality is understood as limited
to penal purposes or as independent of those purposes. The
arguments most frequently raised against a proportionality
requirement for criminal sentences focus on institutional competence,
legislative prerogative, and the difficulty of developing an objective
standard.7 To a significant degree, these arguments depend on the
assumption that proportionality is inextricably linked to a theory of
penal purpose. For example, concerns about institutional competence
have more force if one thinks that proportionality review requires an
inevitably partisan choice among competing penological theories. But
if proportionality review is an attempt to specify the outer limits of
the penal power—not an attempt to direct the legislature’s choices
within the boundaries of that power—the institutional competence
challenge to proportionality review is less persuasive. Of course,
judicially imposed limits on legislative action will sometimes have
countermajoritarian consequences and will thus continue to be
controversial. But courts and scholars have tended to overstate the
degree to which proportionality review requires judges to meddle in
affairs traditionally and properly left to legislative bodies.8
At present, both proportionality concerns and other sentencing
9
issues enjoy considerable judicial and academic attention. The
7. See infra notes 193–215 and accompanying text.
8. For example, the Supreme Court has repeatedly asserted a principle of “legislative
primacy” as a reason for the judiciary to avoid searching proportionality review. See, e.g.,
Harmelin v. Michigan, 501 U.S. 957, 1001 (1991) (Kennedy, J., concurring) (citing “the primacy
of the legislature” as one of four principles commanding judicial deference on questions of
proportionality); Rummel v. Estelle, 445 U.S. 263, 275 (1980) (stating that proportionality
determinations involve a “basic line-drawing process that is pre-eminently the province of the
legislature”).
9. At about the same time that the Supreme Court announced a minimal Eighth
Amendment proportionality guarantee in Ewing v. California, 538 U.S. 11, 21 (2003) (but
before the Supreme Court decisions in Blakely and Booker), the American Law Institute issued
a report on proposed changes to the sentencing provisions of the Model Penal Code (MPC).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
267
Supreme Court’s retreat from proportionality review of prison
sentences has coincided with a more searching proportionality review
10
11
of death sentences and of civil punitive damage awards. Several
12
recent articles discuss this apparent inconsistency. Most of the recent
commentators on proportionality in the criminal sentencing context
argue, as I do, for a constitutional proportionality requirement that is
broader than the minimalist standard that the Court currently
MODEL PENAL CODE 1–6 (Sentencing Report 2003) [hereinafter MPC SENTENCING REPORT].
This report criticized the existing MPC sentencing provisions for omitting a proportionality
requirement. Id. at 34–36. The Eighth Amendment proportionality debates seem to have been
overshadowed, at least temporarily, by the controversy over the Court’s recent cases requiring
all facts determinative of sentence severity to be found by juries (or admitted by defendants)
rather than found by judges. I argue that although these jury sentencing decisions do not use the
language of proportionality, the Court’s concern with marginal increases in the severity of
sentences suggests an implicit assumption that sentences will be proportionate to offenses. See
infra Part III. The commentary on Blakely is already voluminous, and Booker is likely to ensure
that scholars focus on Sixth Amendment sentencing issues for some time. The academy’s
reaction to Blakely is evident in the titles of some of the earliest articles to address the decision.
See, e.g., Douglas A. Berman, The Blakely Earthquake and Its Aftershocks, 16 FED. SENT’G
REP. 307 (2004); Frank O. Bowman, Train Wreck? Or Can the Federal Sentencing System Be
Saved? A Plea for Rapid Reversal of Blakely v. Washington, 41 AM. CRIM. L. REV. 217 (2004).
10. See Roper v. Simmons, 125 S. Ct. 1183, 1200 (2005) (holding that the death penalty is
disproportionate punishment for offenders under eighteen years of age at the time of the
offense); Atkins v. Virginia, 536 U.S. 304, 317–19 (2002) (holding that the death penalty is an
excessive punishment for mentally retarded offenders).
11. See State Farm Mut. Auto. Ins. Co. v. Campbell, 538 U.S. 408, 429 (2003) (reversing a
judgment for $145 million in punitive damages on the grounds that the award “was neither
reasonable nor proportionate to the wrong committed”); BMW of N. Am., Inc. v. Gore, 517
U.S. 559, 585–86 (1996) (reversing a judgment for $2 million in punitive damages as “grossly
excessive”).
12. See, e.g., Erwin Chemerinsky, The Constitution and Punishment, 56 STAN. L. REV. 1049,
1070–79 (2004) (arguing for both procedural consistency and a unified approach to
proportionality for all forms of punishment); Richard S. Frase, Excessive Prison Sentences,
Punishment Goals, and the Eighth Amendment: “Proportionality” Relative to What?, 89 MINN.
L. REV. 571, 607–09 (2005) (arguing that the “retributive proportionality” and “utilitarian
proportionality” limits placed on punitive damage awards by the Supreme Court could be used
similarly to define constitutional limits on prison sentences); Pamela S. Karlan, “Pricking the
Lines”: The Due Process Clause, Punitive Damages, and Criminal Punishment, 88 MINN. L.
REV. 880, 882–83 (2004) (arguing that a proportionality determination is inherently subjective
and that the Supreme Court gives more decisionmaking authority to juries in criminal
sentencing than punitive damage awards because criminal punishments are institutionally
limited by the role of the executive in a criminal proceeding); Youngjae Lee, The Constitutional
Right Against Excessive Punishment, 91 VA. L. REV. 677, 699 n.102 (2005) (noting the tension
between the Court’s disparate use of proportionality in the Eighth Amendment cases and the
Due Process cases); Rachel A. Van Cleave, “Death Is Different,” Is Money Different? Criminal
Punishments, Forfeitures, and Punitive Damages—Shifting Constitutional Paradigms for
Assessing Proportionality, 12 S. CAL. INTERDISC. L.J. 217, 272–78 (2003) (arguing that the
Supreme Court should apply consistent criteria to all forms of punishment, regardless of
whether that punishment is a deprivation of life, liberty, or property).
032006 02_RISTROPH.DOC
268
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
13
applies. But these advocates of proportionality review have
damaged their own cause by linking proportionality to particular
penological theories.14 By divorcing proportionality from
considerations of penal purpose, I offer a response to the concern
about institutional competence. A central challenge of this Article is
to find a way to specify the limits of the penal power without adopting
a particular theory of the justification of that power.
An argument for a not-specifically-penological principle of
proportionality might run thus: The power to incarcerate or execute is
not absolute. It is always limited by general political principles—such
as respect for individual liberty and equality—that stand independent
of penal purposes. More specifically, each individual’s interest in
liberty means that restrictions on liberty must be proportionate to the
conduct that allegedly justifies the restriction. The power to punish15
does not even exist absent certain conditions: this power comes into
being only after an individual engages in specific conduct that has
been criminalized.16 When the power to punish comes into being, it is
13. Most arguments for proportionality review are variants of the call for the
“constitutionalization” of substantive criminal law, a call famously made by Henry M. Hart
almost 50 years ago and subsequently echoed by many others. See Henry M. Hart, Jr., The Aims
of the Criminal Law, 23 LAW & CONTEMP. PROBS. 401, 409–11 (1958); Herbert L. Packer, The
Aims of the Criminal Law Revisited: A Plea for a New Look at “Substantive Due Process,” 44 S.
CAL. L. REV. 490, 494–95 (1970) (arguing for the subjection of criminal sanctions to a rational
basis test that focuses on economic costs and moral arguments, particularly for so-called
victimless crimes); William J. Stuntz, Substance, Process, and the Civil-Criminal Line, 7 J.
CONTEMP. LEGAL ISSUES 1, 29–38 (1996) (arguing that constitutional limits on criminal
substance, such as a rule against strict liability or a doctrine of desuetude, would prevent the
manipulation, and ultimately the impotence, of constitutional limits on criminal procedure).
These calls have gone unheeded for the most part. General constitutional restrictions on the
substantive criminal law are, like proportionality review, often rejected by courts as improper
judicial meddling in matters of legislative prerogative. For an overview of the failed quest for a
substantive constitutional criminal law, see Louis D. Bilionis, Process, the Constitution, and
Substantive Criminal Law, 96 MICH. L. REV. 1269, 1272–99 (1998).
14. See, e.g., Frase, supra note 12, at 588–596 (finding proportionality requirements in both
retributive and utilitarian theories of punishment); Lee, supra note 12, at 704–709 (arguing that
the prohibition of cruel and unusual punishments establishes a retributive proportionality
principle as a side constraint on the power to punish).
15. The power to punish is composed of subsidiary powers exercised by various state
actors. The power to authorize punishment in the first instance—the power to define activity as
criminal—is distinct from the power to impose punishment. The first power is exercised by the
legislature against all those subject to the criminal laws. The second power is exercised by the
judiciary and then by the executive, and it is exercised only against those individuals who are
convicted and sentenced, and who actually serve their sentences.
16. This principle underpins the void-for-vagueness doctrine of Papachristou v. City of
Jacksonville, 405 U.S. 156 (1972). The Papachristou Court explicitly rejected legislative
authorization of a generalized power to punish: “It would certainly be dangerous if the
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
269
not unlimited. What I will call liberty-interest proportionality is thus
based on a claim that the scope of the penal power (and of the
subsidiary powers exercised by the legislative, judicial, and executive
branches) bears some relation to the conduct that gives rise to this
power. This understanding of proportionality in criminal sentencing is
parallel to proportionality requirements in a variety of other contexts.
Proportionality is often invoked to limit an exercise of state power
according to the scope of the conduct or injury that the state seeks to
address.17
Similarly, each individual in a liberal democracy has an interest
in equal treatment before the law. Equality-interest proportionality is
the requirement that similarly situated defendants convicted of
similar crimes receive similar sentences. This type of proportionality
seems to be more widely accepted than liberty-interest
proportionality; it was equality-interest proportionality that
motivated the federal government and many states to adopt the
sentencing guidelines approaches that Blakely and Booker have now
rejected.18 Equality-interest proportionality is perhaps more
accurately called uniformity,19 but to follow common practice, I will
refer to uniformity as a variant of proportionality.20
Part I reviews a number of theoretical accounts of
proportionality in punishment to distinguish between specifically
penological arguments for proportionality and more general political
arguments for proportionality. The Supreme Court uses the term
penological to refer to theoretical accounts of punishment that
legislature could set a net large enough to catch all possible offenders, and leave it to the courts
to step inside and say who could be rightfully detained, and who should be set at large.” Id. at
165 (quoting United States v. Reese, 92 U.S. 214, 221 (1875)).
17. See infra Part II.
18. See, e.g., Blakely v. Washington, 124 S. Ct. 2531, 2540 (2004) (citing Wash. Rev. Code
9.94A (2000)) (noting that Washington’s sentencing guidelines were based on concerns about
“proportionality to the gravity of the offense and parity among defendants”); see also Stephen
Breyer, The Federal Sentencing Guidelines and the Key Compromises Upon Which They Rest, 17
HOFSTRA L. REV. 1, 4–5 (1988) (noting Congressional intent to reduce sentencing disparities
among similarly situated defendants).
19. One of the architects of the Federal Sentencing Guidelines, then Circuit Judge Stephen
Breyer, described “uniformity and proportionality” as central (but competing) goals of a
sentencing system. “Uniformity essentially means treating similar cases alike,” whereas
proportionality, as Breyer described it, requires “treat[ing] different cases differently.” Breyer,
supra note 18, at 13. In other words, uniformity requires that all similarly situated murderers get
similar sentences, and proportionality requires that murders and thefts be punished differently.
20. See infra note 162 for a discussion of Pulley v. Harris, 465 U.S. 37 (1984), and
“comparative proportionality review.”
032006 02_RISTROPH.DOC
270
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
include claims about the appropriate purpose of punishment; I use
the word in the same way. Political arguments for proportionality
consider the structure and justification of the entire political system,
21
not just the specific practice of punishment. Political proportionality
is, in essence, the consequence of the limits of penological theory—
given society’s commitments to political ideals unrelated to the
problem of crime, any justification of punishment will go only so far.
Part II shifts the focus from theory to constitutional doctrine to
examine political proportionality in an array of constitutional
contexts. With respect to incarceration and capital punishment, the
Supreme Court has created only minimal requirements of libertyinterest proportionality, but has been somewhat more inclined to
require equality-interest proportionality. When neither imprisonment
nor execution is at stake, the Court has been considerably more
supportive of proportionality requirements. The Court has been
especially solicitous to what might be called property-interest
proportionality—limitations on state power to take property from
civil or criminal litigants.
After examining the conceptual basis of political proportionality
in Part I, and the doctrinal support for political proportionality in Part
II, this Article turns, in Part III, to consider how one might
implement political proportionality in the criminal sentencing
context. One source of inspiration may be the jurisprudence of
constitutional criminal procedure, in which limitations on the penal
power are not dependent upon the state’s purpose in punishing.
Instead, courts have viewed the law of criminal procedure as an
external limitation on the power to punish. Proportionality, I argue, is
best understood as a similar external limitation. To implement a
proportionality requirement without reference to penological
purpose, one might look to Booker and other recent Sixth
Amendment sentencing decisions that insist on a close link between
proven criminal conduct and the imposed sentence. Conduct is much
better suited than penology to serve as the determinant of the outer
limits of the power to punish.
21. As explained in more detail below, this proportionality principle is political in the sense
that it relates to the distribution of government power within a political system. See infra pp.
284–87. It is not “political” in the way that many legal scholars like to use that adjective:
majoritarian and/or partisan (as in “the political branches”).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
271
I. THE LIMITS OF PENOLOGICAL THEORY
Proportionality is often conflated with the principle of just
deserts.22 For example, Justice Scalia has argued that the Eighth
Amendment contains no “guarantee against disproportionate
sentences”23 because proportionality “is inherently a concept tied to
24
the penological goal of retribution,” and the Constitution “does not
mandate adoption of any one penological theory.”25 It is certainly the
case that the requirement that punishments be proportional to
offenses has been articulated most often in the context of particular
penological theories. It is not the case that proportionality is
defended more extensively by retributive theories than by any other
penal theory; nor is it the case that proportionality depends on any
particular view of the purpose of punishment. In this Part, I examine
arguments for proportionality as they appear in a range of theories of
punishment: deterrence, incapacitation, rehabilitation, and
retribution. Many philosophical accounts of the proportionality
principle entail two separate requirements: a floor, or a minimum
amount of punishment for a given crime, and a ceiling, or a maximum
punishment for the crime. Although the various arguments for a floor
are linked to the particular penal purpose embraced by the theory,
the arguments for a ceiling are often based on broader political
principles of utility, individual rights, or human dignity. Accordingly,
proportionality as a ceiling on punishment can and should be
understood as a limitation on government power that is independent
of any specific penological theory. I refer to this nonpenological
principle as political proportionality.26 Political proportionality is
compatible with a range of penological theories, but it is not
dependent on any one of them.
22. See, e.g., Hyman Gross, Proportional Punishment and Justifiable Sentences, in
SENTENCING 272 (Hyman Gross & Andrew von Hirsch eds., 1981) (“The principle of
proportion between crime and punishment is a principle of just desert that serves as the
foundation of every criminal sentence that is justifiable.”).
23. Ewing v. California, 538 U.S. 11, 31 (2003) (Scalia, J., concurring) (quoting Harmelin v.
Michigan, 501 U.S. 957, 985 (1991) (Scalia, J., concurring)).
24. Id.
25. Id. (quoting id. at 25 (plurality opinion) (quoting Harmelin, 501 U.S. at 999 (Kennedy,
J., concurring in part and concurring in the judgment))).
26. See infra text accompanying note 81.
032006 02_RISTROPH.DOC
272
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
A. Utilitarian Proportionality
The most sustained and detailed arguments for proportional
punishments come not from retributive theorists, but from their
philosophical adversaries: advocates of utilitarian theories of
punishment.
Jeremy Bentham is known as the father of utilitarianism and as a
leading theorist of punishment, but on both counts he owes much to
Cesare Beccaria. The Italian Beccaria published his best-known work,
Of Crimes and Punishments, when Bentham was only sixteen (though
already graduating from Oxford).27 In that book, Beccaria argues that
government and justice must be based on the principle of utility—the
28
greatest happiness for the greatest number. Though this phrase is
commonly attributed to Bentham as the “fundamental axiom” of
utilitarianism, it is a translation of Beccaria’s “la massima felicita
divisa nel maggior numero.”29 The familiar invocation of “pleasure
and pain” as the “moving powers of sentient beings” was also used
30
first by Beccaria.
Utilitarianism is, of course, a theory of politics that reaches far
beyond the practice of punishment. Though Beccaria’s short treatise
is focused primarily on crime and punishment, it articulates broad
political principles that apply to the structure of government more
generally. Penal institutions and practices are only a subset of the
institutions and practices that make up a political system. Beccaria
clearly considers punishment as part of a larger political context, and
consequently, he offers two different kinds of arguments for
proportional punishments. He sometimes argues for proportional
punishments on specifically penological grounds—on the basis of the
special purpose of punishment—but more often, he argues for
proportionality on broader political grounds.
27. CESARE BECCARIA, OF CRIMES AND PUNISHMENTS (Jane Grigson trans., Marsilio
Publishers 1996) (1764). Both Beccaria and Bentham were prodigies of sorts; Beccaria was only
twenty-five when he wrote Of Crimes and Punishments, which was first published in 1764.
Jeremy Bentham was admitted to the bar at age twenty-one in 1769 and allegedly began to read
Beccaria at about that time. See Ross Harrison, Introduction to JEREMY BENTHAM, A
FRAGMENT ON GOVERNMENT x (J.H. Burns & H.L.A. Hart eds., Cambridge Univ. Press 1988)
(1776); see also Principal events in Bentham’s life, in BENTHAM, A FRAGMENT, supra, at xxiv.
28. See BECCARIA, supra note 27, at 77.
29. Harrison, supra note 27, at vi, xiv.
30. BECCARIA, supra note 27, at 74.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
273
Beccaria rejects retributive theories of punishment and argues
for a system of publicly engineered penalties that would deter crimes
and generate greater utility:
[T]he purpose of punishments is neither to torture and afflict a
sentient creature nor undo a crime already done. . . . The aim, then,
of punishment can only be to prevent the criminal committing new
crimes against his countrymen, and to keep others from doing
likewise. Punishments, therefore, and the method of inflicting them,
should be chosen in due proportion to the crime so as to make the
most efficacious and lasting impression on the minds of men, and the
31
least painful impressions on the body of the criminal.
Punishment must be severe enough to create a “lasting impression on
the minds of men,” but not so severe that it becomes torture. In other
words, Beccaria’s proportionality requirement can be conceptualized
as two subsidiary requirements, each with a separate justification. The
minimum punishment, or the floor, is determined by deterrence
concerns. But the ceiling on punishment is an antitorture principle
that is based on humanist concerns rather than calculations of
torture’s efficacy as a deterrent.32
Beccaria’s objections to cruelty, and thus his call for a ceiling on
the sanction, stem from general principles of human dignity and
33
individual freedom. Put differently, utility maximization is not
Beccaria’s only concern: “[T]hough a punishment may have a good
31. BECCARIA, supra note 27, at 49 (emphasis added); see also id. at 74 (“Therefore the
obstacles to crimes should be stronger according to the degree in which those crimes are
contrary to the public good, and the degree of incentives which causes them.”). But see Morris
B. Hoffman, The Case for Jury Sentencing, 52 DUKE L.J. 951, 996 n.164 (2003) (“Cesare
Beccaria is generally credited with the first rigorous exposition of proportionality as a theorem
of retribution.”). Although Beccaria does not use the language of retribution, Judge Hoffman
seems to assume that because Beccaria insists on guilt, he is necessarily a retributivist. After all,
a typical complaint lodged against utilitarian theories of punishment is that in some
circumstances, they would permit a state to punish an innocent person as a scapegoat. But guilt
is not an exclusively retributive notion. There is a difference between the claim that the guilty
must be punished and the claim that only the guilty can be punished. Retributivists typically
make both claims, but many nonretributivists make the latter claim. See also infra Part III for a
discussion of constitutional guilt.
32. Beccaria does argue at one point that torture is ineffective, BECCARIA, supra note 27,
at 35–40, but he makes this claim to criticize the practice of investigative torture, not to
denounce excessive sentences.
33. For Beccaria, human dignity, individual freedom, and utilitarianism are not mutually
exclusive. He says that if he can demonstrate that the death penalty is “neither useful nor
necessary,” he will “have won the cause of humanity.” Id. at 53. I emphasize his concerns for
individual freedom and human dignity here only to show that his objections to excessive
punishments are not strictly deterrence-based.
032006 02_RISTROPH.DOC
274
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
result, it is not on that account always just; to be just a punishment
34
must be necessary . . . .” Of Crimes and Punishments is filled with
references to the cruelty of extant penal practices.35 Cruelty is
“useless,” but beyond its disutility, it provokes in Beccaria “horror
36
and disgust.” Beccaria remains aware of a truth often forgotten by
those who view punishment as a moral and constructive activity:
every act of punishment, justified or not, is an interference with
individual freedom. “Everything beyond” what is necessary to deter
“is superfluous, and therefore tyrannical.”37 The ceiling on acceptable
punishment is a restriction on excessive, tyrannical government
power. In fact, Beccaria concludes his essay with the claim that it is
(in part) proportionality that distinguishes punishment from violence:
“In order that punishment should never be an act of violence
committed by one or many against a private citizen, it is essential that
it be . . . as little as the circumstances will allow, proportionate to the
crime, and established by law.”38
In contrast to these general political claims, some of Beccaria’s
arguments for proportionality depend on his particular account of the
purpose of punishment. For example, he argues that different crimes
must be punished differently, or would-be offenders will have no
incentive to limit themselves to the least harmful offenses. This
particular demand for proportionality is based on specific penological
assumptions: that punishment aims to deter, and that altering
incentives will change the behavior of would-be wrongdoers. Thus,
adumbrating yet another of Bentham’s themes, Beccaria argues that
the death penalty is objectionable because it is too crude an
instrument; it cannot be adjusted to fit the particular offense. No wise
government should apply the same penalty to “the man who kills a
34. Id. at 62.
35. See, e.g., id. at 8 (“[V]ery few men have examined and set themselves against the
cruelty of punishments and the irregularity of criminal procedure . . . .”); id. at 29 (“Among the
evident yet time-honored abuses . . . must be counted the custom of leveling secret
accusations.”); id. at 34 (“The torture of an accused man while the case against him is being
prepared is a cruelty consecrated by long usage among the majority of nations . . . .”).
36. Id. at 49.
37. Id. at 50; see also id. at 119 (“In order that punishment should never be an act of
violence . . . it is essential that it be . . . as little as circumstance will allow, proportionate to the
crime, and established by law.”).
38. Id. at 119 (emphasis omitted).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
275
pheasant, the man who murders another man, or the man who
39
falsifies an important document . . . .”
This last argument for proportionality is a specifically penological
argument in the sense that it is based on a claim about the purpose of
punishment. Similarly, Beccaria’s argument for a floor or minimum is
also a penological argument. But most of Beccaria’s arguments for a
ceiling on the permissible sanctions for a given crime are political
arguments that do not require acceptance of a deterrence theory of
punishment.
On the implementation of his proportionality principle, Beccaria
is somewhat vague.
Were geometry adaptable to the infinite and obscure combinations
of the actions of men, doubtless there would be a corresponding
scale of punishments which would descend from the [most severe] to
the lightest . . . . But the wise legislator will be content to indicate
the chief divisions on the scale without upsetting its order and
inflicting punishments of the lowest degree for crimes of the first
40
degree.
Adapting geometry to “the infinite and obscure combinations of
the actions of men” is, of course, Bentham’s delight. With occasional
nods to Beccaria, Bentham devises a sort of moral mathematics both
for proportionality in punishment and for utilitarianism more
generally. Pleasures and pains are the instruments with which the
legislator must work, and these instruments can be measured
according to four primary qualities: intensity, duration, certainty (or
41
uncertainty), and propinquity (or remoteness). It is unnecessary to
address here Bentham’s lengthy instructions on how to use these
39. Id. at 74. Beccaria frequently acknowledges his debt to Montesquieu, see, e.g., id. at 9,
and on this point Montesquieu’s influence seems particularly clear.
It is an essential point that there should be a certain proportion in punishments,
because it is essential that a great crime should be avoided rather than a smaller . . . .
....
In China, those who add murder to robbery are cut in pieces: but not so the others; to
this difference it is owing that though they rob in that country they never murder.
In Russia, where the punishment of robbery and murder is the same, they always
murder.
BARON DE MONTESQUIEU, THE SPIRIT OF THE LAWS 89–90 (Thomas Nugent trans., Hafner
Publ’g 1962) (1748).
40. BECCARIA, supra note 27, at 75–76.
41. JEREMY BENTHAM, THE PRINCIPLES OF MORALS AND LEGISLATION 29 (Prometheus
Books 1988) (1781).
032006 02_RISTROPH.DOC
276
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
42
qualities (as well as two subsidiary qualities, fecundity and purity ) to
conduct a utilitarian analysis of any human act. Instead, consider
Bentham’s rules for allocating the pain of punishment to individual
offenses. Bentham is even more explicit than Beccaria in his
description of proportionality as a number of separate requirements,
some mandating a minimum punishment, others determining a
maximum punishment.
The first rule of proportionality (for which Bentham cites
Beccaria) requires that “[t]he value of the punishment must not be
less in any case than what is sufficient to outweigh that of the profit of
43
the offence.” Bentham explains further that when assessing
proportionality, it is necessary to consider the “value” of punishment
rather than its “quantity”: “For the word quantity will not properly
include the circumstances either of certainty or proximity . . . .”44 And
the quantity of punishment is itself composed of two separate factors:
45
intensity and duration.
Bentham identifies several other rules: greater “mischiefs”
should incur greater penalties; punishments should be incrementally
increased to give would-be mischief doers an incentive to do as little
mischief as possible; punishment ought never to be “more than what
is necessary” to deter in accordance with the rules previously
specified; and punishments for similar crimes should be roughly
similar, taking into account the particular circumstances of the
46
individual offender. Most of these rules, Bentham recognizes, “mark
out the limits . . . below which a punishment ought not to be
47
diminished.” Only the rule that punishments ought not to exceed
“what is necessary” specifies “the limits on the side of increase.”48 Or,
to use the terminology I invoked earlier, there are reasons for a
42. Id. at 30–32.
43. Id. at 179.
44. Id. at 183.
45. Id. at 187. These various elements of the value of punishment suggest a weakness in any
proportionality analysis that looks only at the length of a prison term—and not at the prison
conditions and other variables that can affect the severity of a particular individual’s
punishment.
46. Id. at 182. Different individuals will experience similar penalties as different degrees of
hardship, Bentham argues. “The same nominal punishment is not, for different individuals, the
same real punishment.” JEREMY BENTHAM, PRINCIPLES OF PENAL LAW (1771), reprinted in 1
THE WORKS OF JEREMY BENTHAM 401 (John Bowring ed., 1843) [hereinafter BENTHAM,
PRINCIPLES OF PENAL LAW].
47. BENTHAM, supra note 41, at 182.
48. Id.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
277
punishment floor, and there are (possibly different) reasons for a
49
ceiling.
As with Beccaria, Bentham’s arguments for minimum
punishment are linked to the specific purpose of punishment
(deterrence), but his insistence on a maximum permissible
punishment is often based on a broader political claim about pain as
50
disutility. “Punishment, whatever shape it may assume, is an evil.”
Error on the minimum side “is least likely to occur,” but overly
severe punishment “is that to which legislators and men in general are
naturally inclined.”51 The maximum limit on punishment, a rule to be
enforced in most cases by judges in review of legislatures,52 is the
proportionality rule for which “we should take the most
53
precautions.”
Proportionality is thus well-established within deterrence
theories of punishment. But two other oft-cited utilitarian purposes of
punishment, incapacitation and rehabilitation, seem to lack any
54
internal proportionality principle. In fact, the American Law
Institute’s recent Sentencing Report blames the separate pursuits of
incapacitation and rehabilitation for disproportionately severe
criminal sentences imposed in the United States today.55 If sentencing
policymakers invoked utilitarian principles consistently, however, the
same political arguments for an upper limit on punishment advanced
by Beccaria and Bentham would apply to a penal system that seeks to
incapacitate or rehabilitate offenders.
49. See BENTHAM, PRINCIPLES OF PENAL LAW, supra note 46, at 399 (“Punishments may
be too small or too great; and there are reasons for not making them too small, as well as for not
making them too great.”).
50. Id. at 390.
51. Id. at 401. Bentham here suggests a possible response to those who would resist
proportionality review on procedural justice grounds—those who argue that criminal laws
produced by majoritarian political procedures must not be second-guessed by activist judges.
Criminal sentencing may be an area in which democratic processes are likely to produce unjust
results. See infra Part III.
52. BENTHAM, supra note 41, at 182.
53. BENTHAM, PRINCIPLES OF PENAL LAW, supra note 46, at 401.
54. Rehabilitation and incapacitation are most frequently justified in consequentialist,
utilitarian terms: it serves the greater good to rehabilitate or incapacitate offenders. But see
Franklin E. Zimring, Principles of Criminal Sentencing, Plain and Fancy, 82 NW. U. L. REV. 73,
75 (1987) (suggesting that in some contexts, such as juvenile justice, rehabilitation may be a
moral obligation rather than simply a utilitarian aim).
55. See MPC SENTENCING REPORT, supra note 9, at 35 (“[I]t is difficult to place a ceiling
upon the goal [of] general incapacitation in the absence of a limiting principle derived from
retributive theory . . . .”).
032006 02_RISTROPH.DOC
278
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
Incapacitation is simply a way to prevent future crimes—like
specific deterrence, it is targeted at the individual offender, but unlike
a deterrence strategy, it tries to make crimes physically impossible
56
rather than consequentially unattractive. To the extent that
contemporary criminologists, political leaders, or legal scholars
advocate punishment for the sake of incapacitation, they do so
subject to the limitations imposed by other political values, including
other utilitarian values. The surest form of incapacitation is death,
and the surest way to incapacitate every convicted offender would be
to impose capital punishment for all crimes. Almost no one advocates
such an approach, for incapacitation is hardly society’s only
consideration.57 Even for strict utilitarians, factors such as the high
social costs of capital punishment and mass incarceration necessitate
limits to the pursuit of incapacitation. And of course liberal
utilitarians—those who would pursue general utilitarian aims only
insofar as they are consistent with categorical liberal principles—
would also insist on proportionality as an external limitation on the
power to incapacitate.
Rehabilitation, to the extent that it is still a viable penal theory,58
is also limited by broader political concerns that would impose a limit
on the maximum punishment. Contemporary proponents of
rehabilitative punishments insist that proportionality principles
should curtail the extent to which the state can restrict liberty in its
efforts to rehabilitate.59 Additionally, many who advocate
56. See, e.g., BENTHAM, PRINCIPLES OF PENAL LAW, supra note 46, at 367 (noting that
incapacitation through “physical restraint” is but one method of preventing “the recurrence of
similar offences”).
57. In very recent history, the Bush administration asserted an all-but-absolute right to
incapacitate in certain contexts. The administration argued that it could detain suspected
terrorists and “enemy combatants” indefinitely in the interests of public safety. See Hamdi v.
Rumsfeld, 542 U.S. 507, 519–21 (2004) (rejecting the government’s argument that national
security interests precluded judicial review of detention of alleged “enemy combatants”). This
detention is not quite equivalent to penal incapacitation, for many detainees, including Hamdi
at the time that he brought his petition for a writ of habeas corpus, had not been charged, much
less convicted, of criminal offenses. Id. at 510–11. Beyond the context of terrorism, many recent
laws seek to increase the power of the state to incapacitate certain classes of offenders. For a
survey and critique of these laws, see generally Paul H. Robinson, Punishing Dangerousness:
Cloaking Preventive Detention as Criminal Justice, 114 HARV. L. REV. 1429 (2001).
58. See, e.g., Francis A. Allen, The Decline of the Rehabilitative Ideal, in PRINCIPLED
SENTENCING 23, 22–30 (Andrew von Hirsch & Andrew Ashworth eds., Ne. Univ. Press 1992)
(addressing the scope and causes of the “modern decline of penal rehabilitationism”).
59. See, e.g., Edward Rubin, Just Say No to Retribution, 7 BUFF. CRIM. L. REV. 17, 50
(2003) (arguing that proportionality “serves to limit the punishments prescribed by utilitarian or
instrumental theories of punishment, such as deterrence, incapacitation, and rehabilitation”).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
279
rehabilitative punishments emphasize that other penological
principles—usually, retributive requirements of desert—must first be
60
satisfied before the state can exercise its penal power at all.
B. Just Deserts, Lex Talionis, and Other Retributive Ideals
Though utilitarian theorists give the most comprehensive
accounts of proportionality by far—my description of Bentham’s
views of proportionality is a very brief summary of a very long
discussion—it is retributive theorists who are most commonly
associated with the principle of proportional punishments. Arguments
for proportionality in retributive theories are more often specifically
penological arguments than are the utilitarian arguments for
proportionality. Nevertheless, most retributive accounts of
punishment are parts of larger political theories, and as such, they
contain further political arguments for limitations on punishment that
stand independent of the penal purpose of retribution.
Retributive theories typically portray punishment as the
wrongdoer’s “just deserts.” A principle of just deserts can, but need
not, demand proportionality between offense and sanction; what the
principle really demands is a correspondence between desert and
61
sanction. A criminal sentence should be more or less severe in
accordance with the wrongdoer’s culpability. Notably, the general
rule of just deserts does not itself prescribe how “desert” is to be
determined. Desert is an inescapably moral issue, and the
determination of an offender’s just deserts will depend on a number
of moral judgments. One may judge desert by assessing the moral
gravity of the particular offense or may focus upon the individual
characteristics of the particular offender. Thus, just deserts might be
compatible with individualized sentencing: one might say that
defendants who were provoked, or who suffer mental impairments, or
who repent their crimes and cooperate with authorities thereafter,
deserve less severe punishments.
60. See, e.g., David O. Brink, Immaturity, Normative Competence, and Juvenile Transfer:
How (Not) to Punish Minors for Major Crimes, 82 TEX. L. REV. 1555, 1568 (2004) (“The
retributive formula . . . determines the length or severity of punishment, [but] it does not
otherwise tell us how to punish. . . . Provided that we punish all and only the guilty and that our
punishments are proportional to their desert, we should punish in ways designed to rehabilitate
the offender and deter crime.”).
61. See, e.g., Kyron Huigens, Rethinking the Penalty Phase, 32 ARIZ. ST. L.J. 1195, 1203
(2000) (“The question is only whether, roughly speaking, the punishment imposed is accurate
with respect to the person’s desert.”).
032006 02_RISTROPH.DOC
280
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
A second retributive principle that is often associated with
62
proportionality is lex talionis, or the law as retaliation. In common
parlance, lex talionis is called the principle of “an eye for an eye.”63
Although lex talionis can be interpreted to require an element of
proportionality, it is not first and foremost a principle of
proportionality.64 As typically invoked, lex talionis is rather a bundle
of claims about the nature of crime, the proper method of
punishment, and the effect that punishment should have on the
offender. Kant, one of the most prominent advocates of lex talionis,
explains the principle as one of reciprocity and equality.
But what kind and what degree of punishment does public justice
take as its principle and norm? None other than the principle of
equality in the movement of the pointer on the scales of justice . . . .
Thus any undeserved evil which you do to someone else among the
people is an evil done to yourself. If you slander him, you slander
yourself; if you rob him, you rob yourself; if you strike him, you
strike yourself; and if you kill him, you kill yourself. But it should be
understood that only the law of retribution (ius talionis) can
determine exactly what quality and quantity of punishment is
required, and it must do so in court, not within your private
65
judgment.
Of course, Kant does not mean that when you rob someone else, you
literally rob yourself—were that true, state-imposed punishment
would be unnecessary, for each criminal act would automatically
entail its own penalty. Rather, Kant means to highlight what is
objectionable about crime: it fails to adhere to the dictate of the
categorical imperative that one always act in such a way that the
maxim guiding one’s action could be universalized into a law for
66
Punishment
is
essentially
“an
exercise
in
everyone.
universalization”—it demonstrates to the offender “what it would be
62. Lex talionis is compatible with nonretributive theories of punishment as well. See infra
note 68.
63. “And if any mischief follow, then thou shalt give life for life, eye for eye, tooth for
tooth, hand for hand, foot for foot, burning for burning, wound for wound, stripe for stripe.”
Exodus 21:23–25 (King James).
64. See Jeremy Waldron, Lex Talionis, 34 ARIZ. L. REV. 25, 47 n.46 (1992)
(“Proportionality may or may not be a byproduct of the application of lex talionis.”).
65. IMMANUEL KANT, THE METAPHYSICS OF MORALS, reprinted in KANT: POLITICAL
WRITINGS 131, 155 (Hans Reiss ed., H.B. Nisbet trans., 2d ed. 1991).
66. See IMMANUEL KANT, GROUNDING FOR THE METAPHYSICS OF MORALS 30 (James W.
Ellington trans., Hackett Publ’g 1981) (1785) (“Act only according to that maxim whereby you
can at the same time will that it should become a universal law.”).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
281
like for him if everyone” acted according to the same maxim that
67
permitted his crime.
A key aspect of lex talionis as presented by Kant is the similarity
in method between punishment and crime. In other words, “an eye
for an eye, a tooth for a tooth” is a claim that punishments must
match crimes not only in magnitude, but also in method—“burning
68
for burning.” Because the method of punishment shares key features
with the crime, the punishment requires the punished to live by his
own maxim, to experience life under the law that he prescribed for
himself when he committed his offense. To the extent that
proportionality between crime and punishment is an incidental
requirement of Kant’s lex talionis, it is a penological proportionality
requirement and not a political one—proportionality is intrinsically
linked to the purpose of punishment. But it is not clear whether this
specifically penological proportionality requirement is still politically
viable. The insistence that punishments borrow from the methods of
the corresponding offenses has been subjected to considerable
ridicule.69 In any event, this strict version of lex talionis is probably not
67. Waldron, supra note 64, at 29.
68. See supra note 63. Note that lex talionis is compatible with utilitarian theories of
punishment. For example, Bentham suggest that “offenses against honor” might require
punishments in kind: “For an insult offered to a woman, the man might be muffled up in the
headdress of a woman, and the like insult might be inflicted on him by the hand of a woman.”
BENTHAM, PRINCIPLES OF PENAL LAW, supra note 46, at 381. Such methods are designed “to
transfer to the insolent offender the contempt which he wished to fix upon the innocent.” Id.
69. The ridicule is often initiated with the question whether the state should impose rape as
punishment for rapists. Lex talionis can be interpreted in such a way as to exclude rape for
rapists, but only at considerable cost to the integrity of the principle. Professor Jeremy Waldron
has argued that lex talionis requires only that “the act of punishment be the same as the act that
constituted the offense.” Waldron, supra note 64, at 32. This requirement can be satisfied,
Waldron argues, as long as the punishment shares the “wrong-making characteristics” of the
crime. Id. at 37. The wrong-making characteristics of the crime can be defined at a fairly
abstract level. So, for example, stealing is wrong because it renders property rights insecure and
produces economic uncertainty. Lex talionis could be satisfied, Waldron suggests, by giving a
thief “a taste of economic uncertainty in other ways: for example, sentencing him to community
service on days determined arbitrarily, unpredictably and at the last minute by a probation
officer.” Id. at 44. This version of lex talionis is certainly consistent with Kant’s account. Kant
finds lex talionis satisfied if “a high-ranking official convicted of violence” were sentenced to
make an apology and endure “painful solitary confinement.” KANT, supra note 65, at 156.
“[A]part from the resultant discomfort, the perpetrator’s vanity would also be painfully
affected, and this humiliation would provide an appropriate repayment of like with like.” Id.
Notwithstanding this broad interpretation of lex talionis, it seems to me that the principle is of
little use given the vast range of criminalized activities and the relative uniformity of criminal
sanctions. One must generalize “wrong-making” to an almost meaningless abstraction in order
to find incarceration, fines, and probation sufficient to capture the “wrong-making
032006 02_RISTROPH.DOC
282
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
possible in a penal system that relies almost exclusively on a few
standard sanctions: fines, community supervision, and imprisonment.
Does Kant also offer a political argument for proportionality?
He certainly articulates arguments for limits on punishment that are
based on broad principles of justice unrelated to penal purpose.
Wrongdoers must be punished as much as they deserve, but no more
70
than they deserve. In other words, the fact that an individual has
violated the categorical imperative and committed a crime does not
give the sovereign absolute power over that individual. Even just
punishment is an infliction of pain, and just punishment is just only
insofar as it corresponds to the moral desert of the punished.
Anything beyond that is a violation of the individual’s “inherent
personality” that fails to treat him as an end in himself as required by
the categorical imperative.71
Other retributivists have developed similar arguments,
explaining that desert serves as a “limiting principle”72 or as a “side
73
constraint.” One could play with the word “just” in “just deserts”—
although the phrase is typically used to mean “the deserts that are in
characteristics” of such diverse crimes as drug possession, financial fraud, physical assault, lying
to federal agents, production of child pornography, and so on.
70. See KANT, supra note 65, at 154–55:
[W]hat kind and what degree of punishment does public justice take as its principle
and norm? None other than the principle of equality in the movement of the pointer
on the scales of justice . . . it should be understood that only the law of retribution (ius
talionis) can determine exactly what quality and quantity of punishment is
required . . . .
71. Id. at 155 (“For a human being can never be manipulated just as a means of realising
someone else’s intentions and is not to be confused with the objects of the law of kind. He is
protected against this by his inherent personality . . . .”); see also Don E. Scheid, Kant’s
Retributivism, 93 ETHICS 262, 272 (1983) (noting that for Kant, the purposes of punishment
must “be pursued in a morally acceptable way, that is, in a way which gives full moral respect to
the persons to whom the penal system is applied”).
72. Norval Morris, Desert as a Limiting Principle, in PRINCIPLED SENTENCING, supra note
58, at 201, 201. For Morris, desert determines both upper limits on punishment, NORVAL
MORRIS, THE FUTURE OF IMPRISONMENT 73 (1974), and, for especially serious offenses, lower
limits, id. at 74. The American Law Institute has proposed a codification of Morris’s “limiting
retributivism” into the sentencing provisions of the Model Penal Code. See MPC SENTENCING
REPORT, supra note 9, at 41 (“[B]road support has been voiced for the theory of limiting
retributivism as the philosophical cornerstone of sentencing decisions under the revised Model
Penal Code.”).
73. See, e.g., Lee, supra note 12, at 704 (“Retributivism under the Eighth Amendment . . .
serves as a side constraint on the socially desirable practice of punishment.”). “Side constraints”
is Robert Nozick’s term for absolute or near-absolute individual rights that constrain the actions
of other individuals and of states. See ROBERT NOZICK, ANARCHY, STATE AND UTOPIA 28–31
(1974) (noting that to understand rights as side constraints means that one cannot violate those
rights in pursuit of other goals).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
283
accordance with justice,” it might also mean “only what is deserved.”
Punishment may extend as far as desert, but no farther. Importantly,
these arguments against punishments that exceed desert are
arguments about human dignity and individual rights, not arguments
about penal purpose. It is desert that justifies punishment at all,
according to these authors, but desert can justify only so much.
Some contemporary accounts of retributivism focus on notions of
equality as much as (or more than) they address desert. In one wellknown account of egalitarian retributivism, Herbert Morris posited
that wrongdoers exempt themselves from the burdens of self-restraint
imposed by the criminal law.74 Punishment must then be imposed to
restore the equal distribution of the law’s burdens. Proportionality is
intrinsic to such egalitarian retributivism: since punishment restores a
just distribution, the scope of punishment must necessarily
correspond to the scope of the inequality of burdens and benefits
created by the offending act.75 Egalitarian retributivism could be said
to contain both a specifically penological proportionality
requirement—one intrinsic to the task of restoring equality—and a
broader political requirement that punishments, qua deprivations, be
no more severe than necessary to achieve the goal of restored
equality.76
74. Herbert Morris, Persons and Punishment, reprinted in SENTENCING, supra note 22, at
95 (“[I]t is just to punish those who have violated the rules and caused the unfair distribution of
benefits and burdens. A person who violates the rules has something others have—the benefits
of the system—but by renouncing what others have assumed, the burdens of self-restraint, he
has acquired an unfair advantage.”). See also WOJCIECH SADURSKI, GIVING DESERT ITS DUE:
SOCIAL JUSTICE AND LEGAL THEORY 225–227 (1985) (explaining that punishment is “a
method of restoring an overall balance of benefits and burdens”).
75. Sadurski suggests that we view retributive justice as “the proportional relations
between inputs and outputs”—the inputs are crimes, the outputs punishments. SADURSKI, supra
note 74, at 221; see also id. at 229 (“[C]riminal law reflects the hierarchy of protected values: the
more precious the value, the bigger the benefit of non-self-restraint acquired by the criminal.
The intuitively just principle that more serious crimes should be punished more heavily is not,
therefore, violated by the proposition about punishment as a restoration of the balance of
benefits and burdens.”).
76. See Eric L. Muller, The Virtue of Mercy in Criminal Sentencing, 24 SETON HALL L.
REV. 288, 296 (1993) (“[T]rue retributivism is, at its core, a deeply egalitarian theory of
punishment.”). Muller addresses proportionality in passing and, like some other commentators,
see supra note 22, seems to assume that it is a concept specific to retributive theories. See
Muller, supra, at 297 (“[T]he idea of proportionality between the seriousness of the offense and
the amount of punishment is central to retributivism. . .”); see also id. at 340 (“The retributivist
[in contrast to the utilitarian] would insist that the punishment bear some sense of proportion to
the . . . nature of the criminal’s wrongdoing.”). Nevertheless, Muller’s arguments against
excessive sentences, and for mercy in some instances, are explicitly based on the “equal inherent
worth” of the offender as a human being. See id. at 296. This rationale for a limitation on
032006 02_RISTROPH.DOC
284
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
Thus, to many retributivists, moral desert provides both a floor
and a ceiling for punishment. To understand why punishment below
the floor is unacceptable, one must look to the argument for
punishment: society needs to make criminals experience the
wrongness of their own acts by inflicting on them harm similar in kind
and degree. But to understand why punishment above the ceiling is
unacceptable, one must look beyond the theory of punishment, to
broader arguments such as the Kantian instruction to respect persons
as ends in themselves.77
C. Political Proportionality
Proportional punishments are required in a wide variety of
penological perspectives, including the mainstream penological
theories typically advanced in American legal and political
78
discourse. But even if a state were to abandon retribution,
sentence severity is not necessarily retributive; presumably, one need not be a retributivist to
believe that all human beings have equal inherent worth.
77. I do not mean to suggest that every retributive theory of punishment can be interpreted
to include a political argument for proportionality. For example, Hegel certainly demands
proportionality in punishment, but his argument seems dependent on his particular account of
the purpose of punishment—and on his unusual claims about the metaphysical effects of crime.
Hegel argues that crime, or wrong, has a continuing presence (“a positive external existence”)
even after the moment the wrong is committed. G. W. F. HEGEL, ELEMENTS OF THE
PHILOSOPHY OF RIGHT 123, § 97 (H. B. Nisbet trans., Cambridge University Press 1991) (italics
in Nisbet’s translation). Punishment serves to negate the crime and restore the world to its prewronged state. See id. § 97, Addition (“The criminal act . . . is itself negative, so that the
punishment is merely the negation of the negation.”) Hegel’s argument requires proportionality
but not the strong form of lex talionis—the scope of the punishment must correspond to the
scope of the injury, but the method of punishment need not correspond to the method of crime.
Hegel, like other retributive theorists, assumes commensurability between diverse injuries, but
Hegel is particularly explicit about this commensurability requirement. He assumes that we can
measure the scope of wrong generated by, say, a physical assault, and choose a corresponding
punishment that inflicts injury of the same scope. “The cancellation [Aufheben] of crime is
retribution in so far as the latter, by its concept, is an infringement of an infringement, and in so
far as crime, by its existence [Dasein], has a determinate qualitative and quantitative magnitude,
so that its negation, as existent, also has a determinant magnitude. But this identity . . . is not an
equality in the specific character of the infringement, but in its character in itself—i.e., in terms
of its value.” Id. at 127, § 101; see generally id. at 121–32, §§ 95–104.
78. Restorative justice, a penological paradigm not previously discussed in this Article,
arguably does not contain a proportionality requirement. Some commentators have suggested
that restorative justice actively resists at least some versions of proportionality in criminal
sentencing. See, e.g., Sara Sun Beale, Still Tough on Crime? Prospects for Restorative Justice in
the United States, 2003 UTAH L. REV. 413, 433 (“Almost all the sentencing guidelines reflected
core principles that are in conflict with the restorative justice movement, namely,
proportionality . . . .”); John Braithwaite, A Future Where Punishment is Marginalized: Realistic
or Utopian?, 46 UCLA L. REV. 1727, 1743 (1999) (noting that the restorative justice model
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
285
deterrence, rehabilitation, and incapacitation in favor of some other
penal purpose, basic principles of liberal democracy would still
impose a proportionality requirement as an upper limit on criminal
sentences. This political proportionality is a consequence not of
penological theory itself, but rather of the limits of penological theory.
No penological theory does (or could) grant a liberal government
79
absolute power over an individual who breaks the law. Principles
that may justify punishment are not the only principles that should
inform choices about penal practices. Alongside retribution, utility
maximization, and other goals that may lead to the imposition of
punishment, basic liberal principles of liberty and equality demand
restrictions on the manner and degree of punishments. A
proportionality requirement is thus a consequence of competing
political principles, and such a requirement is essential to any liberal
account of punishment. In this Section, I explore proportionality as a
component of political theories of state power.
Robert Nozick claims that the fundamental question of political
80
philosophy is “whether there should be any state at all.” The very
identification of this question as fundamental indicates an alreadyexisting ideological commitment: a suspicion of power, a default
assumption that states need to be justified. (Why is the fundamental
question not “whether there should be any liberty at all”?) But this
particular ideological commitment is very widely held—most people
do not start by wondering whether there should be any such thing as
liberty. Liberals are not alone in thinking that coercive state power
needs to be justified. Those who spend (waste?) time pondering such
matters tend to hold the view that state power is at least potentially
an interference with individual liberty. As such, a coercive state must
be scrutinized, argued for, and defended rather than accepted
automatically.
To the anarchist, the initial suspicion of power cannot be
eradicated and no state can be justified. To the totalitarian, not only
power but absolute power is justifiable. Between anarchy and
totalitarianism lies limited government. The kind of proportionality
“involves rejection of a justice that balances the hurt of the crime with proportionately hurtful
punishment”).
79. See supra note 57. It is important to recognize that even the death penalty does not give
a liberal state absolute power over the condemned. Liberal justifications for capital punishment
insist that the condemned retains various rights up to and beyond execution. For example, death
row inmates may not be tortured or abused, and their corpses must be treated with dignity.
80. NOZICK, supra note 73, at 4.
032006 02_RISTROPH.DOC
286
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
requirement that underwrites the American political system and its
legal institutions, including the law of sentencing, is a consequence of
two ideological commitments: the view that state power always needs
to be justified, and the commitment to limited government rather
than to no state or a total state.
To elaborate, it is relatively uncontroversial that (1) state power
must be justified; (2) some (limited) power can be justified; and (3)
absolute power is never justified. From these claims flows a
proportionality principle that is broadly political, not specifically
penological. The state is a set of institutions that exercise a wide
range of powers. Each particular power must be justified, and the
justification must apply to each exercise of the power. For example,
an abstract justification of the power to raise armies does not
necessarily justify full-time and indefinite compulsory military
servitude for all adult citizens. Instead, the principle of political
proportionality holds that the scope of state power should be
proportional to the occasions for the power.
The adjective political deserves some specification here, for the
word is often used by lawyers to mean “majoritarian.” Thus the
judiciary is often contrasted to the “political branches,” and legal
questions are distinguished from political ones. This use of the term
makes “political proportionality” seem ill-suited to serve as the basis
of judicial review of criminal sentences. I use the term “political” as it
is used by “political” scientists—and indeed, as it is used by almost
everyone who is not a lawyer. Political means simply, of a polity. Thus
I mean to contrast political proportionality—a requirement of a
political system—with penological proportionality—a requirement of
a penal theory.
Political here also serves to emphasize that this kind of
proportionality is determined by the members of a particular polity; it
is not dictated by some transcendent authority. To borrow from John
Rawls, this proportionality requirement is “political and not
metaphysical.”81 Again, the fact that a proportionality standard is a
human artifact does not mean that it must be an exclusively
majoritarian project, or that to involve the judiciary is antidemocratic.
When Rawls says that justice is political and not metaphysical, he
81. JOHN RAWLS, POLITICAL LIBERALISM 97 (1993). Rawls argues that a political account
of justice is necessarily distinct from a “comprehensive moral doctrine,” id. at 90–91, and that
justice must be political rather than metaphysical because citizens in a diverse, pluralistic society
will never agree on any single transcendent moral authority, id. at 97.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
287
does not mean that judges have nothing to say about justice. Here is
further reason to resist the use of the term political to exclude the
judiciary: it tends to obscure the importance of separation of powers
and the judiciary’s role in providing checks on the majoritarian
branches.
The relation between proportionality and separation of powers
should be underscored. It is fundamental to theories of limited
government that government bodies cannot be trusted to impose and
observe limitations on their own powers. Accordingly, limitations on
power must come from outside the body that exercises power—from
the people and from other government institutions. So
proportionality as a limit on (for example) legislative power cannot
be left to legislative determination. The phrase “separation of
powers” is used sloppily today, as a shorthand assertion that the
judiciary should just leave the legislature alone, or that Congress
82
should leave the President to do as he pleases. But as the political
scientist Richard Neustadt observed, our constitutional design creates
83
a system of “separated institutions sharing powers.” Separation of
powers does not mean that each branch of government can exercise
its specific powers without limitation, as though individual liberty
were likely to be protected by fierce competition among three tyrants.
To the contrary, separation of powers means that power itself is
separated, divided and distributed across different political
institutions and across the branches. The power to punish is one
governmental power for which such fragmentation is particularly
important.
The proportionality relationship identified here is between
power (such as the power to punish) and the source of the power. In
this regard, it differs from penological proportionality, which focuses
on the relationship between the sanction itself and some determinant
(the offender’s desert, the harm of the crime). In considering whether
82. This spin on “separation of powers” was invoked to explain Harriet Miers’s withdrawal
from consideration for the Supreme Court. After members of Congress demanded that the
White House produce documents related to Miers’s service there, Miers explained that
“[p]rotection of the prerogatives of the Executive Branch and continued pursuit of my
nomination are in tension.” In response, President Bush claimed that “Harriet Miers’s decision
[to withdraw from consideration] demonstrates her deep respect for this essential aspect of the
constitutional separation of powers. . . .” See Timothy Williams, Bush’s Pick for Court
Withdraws Her Name; Miers, Under Attack from Both Parties, Calls Process ‘Burden,’ INT.
HERALD TRIBUNE, October 28, 2005, at 1.
83. RICHARD NEUSTADT, PRESIDENTIAL POWER: THE POLITICS OF LEADERSHIP 29
(1960) (emphasis omitted).
032006 02_RISTROPH.DOC
288
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
political proportionality can serve as the basis of a practical
constitutional standard, it will be important to remember that it is the
outer scope of the power to punish that courts must determine, not
the precise severity of a particular offender’s punishment.
For one example of proportionality as a limitation on
government power, consider John Locke—by many accounts, the
84
venerable godfather of American liberal tradition —on punishment.
According to Locke, the power to punish arises as soon as, but no
sooner than, the law of nature is violated. The power to punish is thus
always limited in scope by the scope of the transgression.
[I]n the State of Nature, one Man comes by A Power over another;
but yet no Absolute or Arbitrary Power, to use a Criminal . . .
according to the passionate heats, or boundless extravagancy of his
own Will, but only to retribute to him, so far as calm reason and
conscience dictates, what is proportionate to his Transgression, which
85
is so much as may serve for Reparation and Restraint.
Note that Locke no sooner announces the power to punish than he
emphasizes that it is not “absolute or arbitrary.” Note also that Locke
uses the terms retribute, reparation, and restraint to justify
punishment, and in fact elsewhere adopts additional justifications,
including deterrence.86 Locke’s proportionality requirement is not
dependent upon a single penological purpose, or even on his hybrid
theory of penal purpose, but upon the natural liberty and equality
that provide the background against which the right to punish
sometimes arises.
In the passage cited above, Locke’s justification for punishment
and his limitations on punishment are matters of natural right; they
do not depend upon the existence of a social contract or civil society.
The social contract ultimately transforms the natural authority to
84. See, e.g., LOUIS HARTZ, THE LIBERAL TRADITION IN AMERICA 5–6 (1963) (arguing
that American democracy “begins with Locke” and “stays with Locke, by virtue of an absolute
and irrational attachment it develops for him”); see also id. at 8 (describing the American South
as “an alien child in a liberal family, tortured and confused, driven to a fantasy life which,
instead of disproving the power of Locke in America, portrays more poignantly than anything
else the tyranny he has had”).
85. JOHN LOCKE, TWO TREATISES OF GOVERNMENT 272, § 8 (Peter Laslett ed.,
Cambridge Univ. Press 1988) (1690) (I have added emphasis, and, for the sake of clarity, I have
removed Locke’s original italics). References to Locke herein include section numbers after the
page number.
86. See, e.g., id. at 274–75, § 12 (“Each Transgression may be punished to that degree, and
with so much Severity as will suffice to make it an ill bargain to the Offender, give him cause to
repent, and terrifie others from doing the like.”).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
289
punish into the state’s political authority to punish, and it transforms
the natural limitations on punishment into political limitations that
restrict the state’s power to punish. In other words, Locke’s
proportionality begins as a matter of natural law. Of course,
liberalism as practiced in contemporary America is not necessarily
based on a shared theory of natural law. Rawls’s “political liberalism”
is at least as good an explanation of why American society adopts the
views of individual rights and legitimate government that it does.87
But for the purposes of my argument, it does not matter whether
American society is more Lockean or more Rawlsian. The key point
is that once one has accepted basic principles of individual freedom
and equality—for whatever philosophical or pragmatic reason—those
principles create a presumption against the exercise of force that any
justification for punishment must overcome. The principle of
proportionality is a requirement that the power to punish reaches
only as far as the injury to be addressed.88
The political nature of the proportionality requirement may be
largely overlooked in legal scholarship because punishment tends to
be theorized as an independent practice, divorced from important
89
background assumptions about liberty, utility, and individual rights.
It is important to resist this tendency, for any theory of punishment
needs to be placed in the context of a larger political theory.
Punishment is the quintessential example of state coercion.
Punishment theory must not limit itself to the particular moral or
economic interests that might be served by such coercion; it needs to
87. See generally RAWLS, supra note 81.
88. Cf. Francis A. Allen, A Matter of Proportion, 4 GREEN BAG 2d 343, 344 (2001) (noting
that the principle of proportionality in punishment used to be “understood in much the same
way as other limitations on governmental powers set forth in the bills of rights of American
federal and state constitutions”).
89. For example, Professor Jeffrie Murphy has explained that two very different accounts
of punishment emerge from Kant’s work depending upon whether one focuses strictly on the
famous passages on punishment in Part I of The Metaphysics of Morals or instead collects
Kant’s observations about punishment across all of his writings. See Jeffrie G. Murphy, Does
Kant Have a Theory of Punishment?, 87 COLUM. L. REV. 509, 518 (1987) (“Is the above account
in fact what we find consistently defended and amplified in the text of the Rechtslehre?
Hardly!”). When one considers all of Kant’s writings other than Part I of The Metaphysics of
Morals, it is clear that for Kant, “justified punishment is a deterrence system functioning to
maintain a system of ordered liberty of action. To set any more morally ambitious goal for
punishment would be to adopt an unacceptable theory of the role of the state and would
represent an attempt to play God . . . .” Id. at 517.
032006 02_RISTROPH.DOC
290
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
draw upon political theory more generally to explain how other
90
political principles circumscribe the use of coercion.
I do not mean to suggest that all theorists of punishment have
overlooked this point. In The Aims of the Criminal Law, Henry M.
Hart emphasizes the larger political context in which punishment
takes place. “A penal code that reflected only a single basic principle
would be a very bad one. Social purposes can never be single or
simple, or held unqualifiedly . . . and an effort to make them so can
91
result only in the sacrifice of other values which also are important.”
Hart argues that background conditions unrelated to penal
purposes—such as individual rights—impose constraints on the way a
state determines guilt and imposes punishments.
Or, in the words of the other Hart:
[I]n relation to any social institution, after stating what general aim
or value its maintenance fosters we should enquire whether there
are any and if so what principles limiting the unqualified pursuit of
that aim or value. Just because the pursuit of any single social aim
always has its restrictive qualifier, our main social institutions always
possess a plurality of features which can only be understood as a
92
compromise between partly discrepant principles.
In the same essay, H.L.A. Hart famously insisted that the “General
Justifying Aim” of punishment was a separate question from
questions about the appropriate distribution of punishment (who to
punish, and how much).93 A proportionality principle guides the
distribution of punishment, and as Hart recognized, such a principle
might be dictated by a retributive account of the general justifying
90. In Professor Murphy’s terms, it is important to distinguish between moral and political
justifications for punishment. Even if one is certain that punishment is morally just, one still
needs an argument for why “the pursuit of these [moral] goals is part of the legitimate business of
the state.” Id. at 510; see also Guyora Binder, Punishment Theory: Moral or Political?, 5 BUFF.
CRIM. L. REV. 321, 371 (2002) (“[T]here is reason to hope that debate about utility and
autonomy in criminal lawmaking will become more productive once it is redefined as a political
debate about institutions rather than a moral debate about the conduct of criminals and
officials.”); Don E. Scheid, Kant’s Retributivism, 93 ETHICS 262, 265–271 (1983) (assessing
whether Kant’s legal justification for punishment is consistent with his moral theories).
91. Hart, supra note 13, at 401.
92. H.L.A. HART, PUNISHMENT AND RESPONSIBILITY: ESSAYS IN THE PHILOSOPHY OF
LAW 10 (1968).
93. See id. at 4, 8–13.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
291
aim of punishment, but it might also be dictated by considerations
94
entirely external to the general justifying aim.
In short, punishment does not take place in a vacuum and it
should not be theorized in a vacuum. Consideration of the larger
political context reveals the limits of penal theory—one sees that
there are some exercises of force that no penal theory can justify.
Political proportionality is a claim about the limits of penal theory in
a liberal state. It is not a theory of punishment, but a theory of the
relationship between state power and individual right. As
described in Part III, individual criminal defendants should be able to
invoke proportionality as a constitutional limitation on their
sentences.
Before elaborating the details of this constitutional right of
criminal defendants, it is worth examining the extent to which
constitutional doctrine has already embraced proportionality as a
principle of limited government. In a number of doctrinal areas
beyond criminal sentences, the Supreme Court has used the concept
of proportionality to limit state power when that power arises and is
exercised in response to specified conduct. In contrast, the Court
has been more reluctant to embrace proportionality in the context
of criminal sentences. But even the criminal sentencing decisions
that first developed the notion of proportionality are better
understood as evaluations of the scope of the power to punish in
light of individual rights—as decisions about political
proportionality—than as vindications or rejections of particular
penological theories.
94. “[I]t is perfectly consistent to assert both that the General Justifying Aim of the
practice of punishment is its beneficial consequences and that the pursuit of this General Aim
should be qualified or restricted out of deference to principles of Distribution which require that
punishment should be only of an offender for an offense. Conversely it does not in the least
follow from the admission of the latter principle of retribution in Distribution that the General
Justifying Aim of punishment is Retribution though of course Retribution in General Aim
entails retribution in Distribution.” Id. at 9. I would not follow Hart in calling a requirement
that punishment be imposed only on the guilty a principle of “retribution in distribution.” A
guilt requirement is justifiable on utilitarian grounds as well as not-necessarily-retributive liberal
grounds (such as the principle of notice).
032006 02_RISTROPH.DOC
292
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
II. PROPORTIONALITY IN CONSTITUTIONAL DOCTRINE
A. Beyond Punishment
For the most part, the United States Constitution itself does not
speak explicitly of “proportionality.”95 But it establishes government
powers and limits those powers. As courts interpret and apply the
Constitution, they often rely on the concept of proportionality to
assess the scope of government powers. To be sure, proportionality as
a constitutional principle is not as frequently invoked in the United
States as it is in some other constitutional democracies. For instance,
courts in Germany and Canada use well-established proportionality
96
tests as tools to limit state power. But even if American courts
invoke proportionality less often than do foreign courts, the
concept—and sometimes the language—of proportionality appears
from time to time in American constitutional decisions. In several
constitutional contexts unrelated to criminal punishment, the extent
of a state power is determined in part by the scope of the problem or
conduct that gave rise to the specific power. A brief survey of these
doctrinal areas demonstrates that a general proportionality
requirement underlies the very concept of limited government and
the application of that concept in American constitutional law.97
95. The U.S. Constitution twice uses the term proportion: Article I, section 9, clause 4
provides that “No capitation, or other direct, Tax shall be laid, unless in Proportion to the
Census or Enumeration herein before directed to be taken.” U.S. CONST. art. I, § 2. Section 2 of
the Fourteenth Amendment specifies the apportionment of representatives to states, and
provides that when male citizens lose their right to vote for reasons other than “participation in
rebellion, or other crime, the basis of representation therein shall be reduced in the proportion
which the number of such male citizens shall bear to the whole number of male citizens twentyone years of age in such state.” U.S. CONST. amend. XIV, § 2.
96. For a description of the proportionality principle under the German constitution, see
David S. Law, Generic Constitutional Law, 89 MINN. L. REV. 652, 689 n.135 (2005) (“Germany
is credited with the invention of proportionality review . . . .”). Under the Canadian Charter of
Rights and Freedoms, intrusions on individual rights may be upheld if the intrusions are
“demonstrably justified in a free and democratic society.” Canadian Charter of Rights and
Freedoms, Canada Act, 1982, ch. 11, §§ 1–4, sched. B (U.K.), as reprinted in R.S.C., No. 44,
(Appendix 1985). Canadian courts apply the “Oakes test” to evaluate intrusions on individual
rights; that test includes an inquiry into whether the means chosen by the government are
proportional to its purpose. Regina v. Oakes, [1986] S.C.R. 103, 114. For a comparative
constitutional law analysis of proportionality, see Vicki C. Jackson, Ambivalent Resistance and
Comparative Constitutionalism: Opening Up the Conversation on “Proportionality,” Rights and
Federalism, 1 U. PA. J. CONST. L. 583, 602–24 (1999).
97. For an overview of “proportionality” tests in several areas of constitutional law, see
K.G. Jan Pillai, Incongruent Disproportionality, 29 HASTINGS CONST. L.Q. 645, 655–93 (2002).
Professor Pillai examines proportionality standards in the contexts of the Fifth Amendment’s
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
293
Proportionality principles—sometimes explicit, sometimes
implicit—underlie the familiar evaluative mechanisms with which
courts adjudicate claims that a state actor has exceeded its power.
When a court requires that a coercive or intrusive state action be
“narrowly tailored” to serve a “compelling state interest,” it enforces
the principle that state power must be proportional to the interest
that allegedly justifies the power. Richard Frase has recently noted
that “narrow tailoring” is a form of proportionality requirement.98
But the notion that state power is limited by proportionality
considerations extends beyond the rare circumstances in which courts
will require “narrow tailoring.” In fact, proportionality is explicitly
invoked in other contexts. For example, to decide whether
“exactions” are permissible under the Fifth Amendment’s Takings
Clause, the Supreme Court applies a “rough proportionality”
99
analysis. Exactions are required dedications of private property for
public use as a condition of permission to develop the property. The
government’s power to exact property from a private landowner must
be proportional in scope to the effects of the proposed development.
“No precise mathematical calculation is required, but the city must
make some sort of individualized determination that the required
dedication is related both in nature and extent to the impact of the
proposed development.”100 Here, a government power (the power to
condition permission to develop on the dedication of private property
to public use) arises when private conduct results in or is likely to
Takings Clause, the Eighth Amendment’s Cruel and Unusual Punishments and Excessive Fines
Clauses, and the Fourteenth Amendment’s Due Process Clause in an attempt to clarify the
meaning of the “congruence and proportionality” standard announced in City of Boerne v.
Flores, 521 U.S. 507, 520 (1997).
98. See Frase, supra note 12, at 618. Frase argues that proportionality underlies an even
broader array of constitutional rules, including not only the Eighth Amendment and the Due
Process Clause but also Fourth and Fifth Amendment exclusionary rules, Fourth Amendment
“reasonableness,” public forum rules under the First Amendment, and balancing under the
Dormant Commerce Clause. See id. at 598–621.
99. City of Monterey v. Del Monte Dunes at Monterey, Ltd., 526 U.S. 687, 702–03 (1999);
Dolan v. City of Tigard, 512 U.S. 374, 391 (1994).
100. Dolan, 512 U.S. at 391. The Dolan Court characterized its decision as a protection of
individual rights against government power:
[S]imply denominating a governmental measure as a “business regulation” does not
immunize it from constitutional challenge on the ground that it violates a provision of
the Bill of Rights . . . . We see no reason why the Takings Clause of the Fifth
Amendment, as much a part of the Bill of Rights as the First Amendment or Fourth
Amendment, should be relegated to the status of a poor relation in these . . .
circumstances.
Id. at 392.
032006 02_RISTROPH.DOC
294
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
result in unwanted social consequences (as when a proposed
development creates an impact on the surrounding area). The scope
of the newly arisen power is limited by the scope of the condition that
101
gave rise to the power.
A particularly controversial proportionality requirement has
been used to determine the scope of Congress’s powers under Section
102
Five of the Fourteenth Amendment. In City of Boerne v. Flores, the
Court found the Religious Freedom Restoration Act to be an
unconstitutional legislative attempt to alter the substantive rights
protected by the Fourteenth Amendment.103 Section Five of the
Fourteenth Amendment gives Congress power “to enforce, by
appropriate legislation, the provisions of this article.” According to
the Court, this enforcement power is remedial and preventive;
legislation adopted under Section Five “must be judged with
reference to the historical experience . . . it reflects.”104 That is, Section
Five enforcement powers arise in conjunction with particular social or
political problems (e.g., discriminatory conditions), and the scope of
those powers is determined by the scope of the relevant problems.
“There must be a congruence and proportionality between the injury
to be prevented or remedied and the means adopted to that end.”105
101. The Del Monte Dunes Court noted that, although “rough proportionality” is a test
specific to the exactions context, “in a general sense concerns for proportionality animate the
Takings Clause.” 526 U.S. at 702. For the purposes of this Article, the more specific
proportionality test used to evaluate exactions is the better illustration of political
proportionality. The exactions test creates a link between private conduct (the conduct of the
property owner) and government power. In the criminal context, there should be a similar link
between the criminal conduct and the power to punish.
102. 521 U.S. 507 (1997).
103. Id. at 536. The Religious Freedom Restoration Act (RFRA), 42 U.S.C. §§ 2000bb–
2000bb-4 (1993), was Congress’s response to Employment Division v. Smith, 494 U.S. 872
(1990), in which the Court held that facially neutral laws that happened to inhibit religious
practices were not subject to heightened scrutiny. Id. at 878–79. The RFRA attempted to
“overrule” Smith; it provided that in order to enforce facially neutral laws that substantially
burdened religious exercise, a state or local government had to demonstrate that the law was the
least restrictive means of serving a compelling government interest. 42 U.S.C. §§ 2000bb–
2000bb-4.
104. City of Boerne, 521 U.S. at 525 (quoting South Carolina v. Katzenbach, 383 U.S. 301,
308 (1966)).
105. Id. at 520. Similarly,
[w]hile preventive rules are sometimes appropriate remedial measures, there must be
a congruence between the means used and the ends to be achieved. The
appropriateness of remedial measures must be considered in light of the evil
presented. Strong measures appropriate to address one harm may be an unwarranted
response to another, lesser one.
Id. at 530 (citation omitted).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
295
The congruence and proportionality standard of City of Boerne is
a useful illustration of the Court’s familiarity with proportionality as a
means of restricting government power in accordance with specific
injuries or problems that give rise to the power. Take, for example,
the different results in two recent cases involving the Americans with
Disabilities Act (“ADA”). In Board of Trustees v. Garrett,106 the
Court found that Congress had exceeded its Section Five powers by
permitting suits against states for employment discrimination under
Title I of the ADA.107 Congress had in fact amassed substantial
evidence of discrimination against disabled persons by states, but
most of the evidence related to “the provision of public services and
public accommodations,” not to employment discrimination.108 The
Court found that the remedy created by Congress (abrogation of state
sovereign immunity) was not proportional to the specific problem to
be addressed (discrimination against the disabled by state
employers).109
In contrast, congressional abrogation of state sovereign immunity
110
in Title II of the ADA was upheld in Tennessee v. Lane. Title II
aims at exactly the kind of discrimination that the Garrett Court
found Congress to have demonstrated most completely:
discrimination against the disabled in the provision of public services
and public accommodations.111 Thus the Lane Court found the same
remedy that had been at issue in Garrett (abrogation of state
sovereign immunity) to be “congruent and proportional” when
applied to a problem that the Court viewed as more substantial
(failure to provide equal access to public services and
accommodations).112 Together, Garrett and Lane illustrate that the
106. 531 U.S. 356 (2001).
107. Id. at 374.
108. Id. at 371 n.7.
109. Id. at 374.
110. 541 U.S. 509 (2004). George Lane, a wheelchair-bound paraplegic and one of the two
plaintiffs, had been called to answer criminal charges in a courtroom on the second floor of a
Tennessee courthouse that had no elevator. On one occasion, Lane crawled up two flights of
stairs to get to the courtroom, but on a second trip to the courthouse for a hearing, Lane refused
to be carried up the stairs or to crawl again. He was arrested and jailed for failure to appear, and
subsequently filed suit against the state under the ADA for failure to provide access to the
courts. Id. at 513–14.
111. 42 U.S.C. §§ 12,131–12,165.
112. Lane, 541 U.S. at 531. Although the Court found Title II of the ADA to pass muster
under the congruence and proportionality test, other doctrinal considerations may also have
contributed to the different outcomes in Garrett and Lane. The Lane Court noted that Title II
was not only a prohibition of “irrational disability discrimination” but also an effort “to enforce
032006 02_RISTROPH.DOC
296
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
Court is familiar with, and willing to employ, a broad proportionality
principle that determines the scope of legislative power with
113
reference to the specific problem that gives rise to the power.
In drawing a comparison between a political proportionality
requirement for criminal sentences and the “congruence and
proportionality” test of City of Boerne, I do not mean to endorse that
particular doctrinal development in the Court’s equal protection
jurisprudence. The test was initially used to strike down several
antidiscrimination laws one after the other114 and has been the target
of considerable scholarly critique.115 In many ways, a proportionality
requirement may be both easier to implement and more useful in the
context of criminal sentences than in the context of federal
antidiscrimination efforts. Here, the reference to City of Boerne and
its progeny is merely a demonstration that nonpenological
proportionality is already well-established in constitutional doctrine.
a variety of other basic constitutional guarantees, infringements of which are subject to more
searching judicial review.” Id. at 522–23.
113. Justice Scalia joined the City of Boerne majority in its original statement of the
congruence and proportionality test, and he subsequently voted with the majority in every case
in which the City of Boerne standard was used to strike down antidiscrimination legislation. See
Board of Trustees v. Garrett, 531 U.S. 356, 372–74 (2001) (applying City of Boerne and finding
that Title I of the Americans with Disabilities Act was not a valid exercise of Congress’s Section
Five enforcement powers); United States v. Morrison, 529 U.S. 598, 619–20 (2000) (applying
City of Boerne and finding that the Violence Against Women Act was not a valid exercise of
Congress’s Section Five enforcement powers); Kimel v. Florida Bd. of Regents, 528 U.S. 62, 82–
83 (2000) (applying City of Boerne to strike down the Age Discrimination in Employment Act
as applied against state and local governments). When the Family and Medical Leave Act of
1993, 29 U.S.C. §§ 2601–2654 (2000), was sustained under the congruence and proportionality
test, Justice Scalia dissented. Nev. Dept. of Human Res. v. Hibbs, 538 U.S. 721, 741 (2003)
(Scalia, J., dissenting). Justice Scalia dissented again in Tennessee v. Lane, and went so far as to
reject the congruence and proportionality test outright: “I yield to the lessons of experience.
The ‘congruence and proportionality’ standard, like all such flabby tests, is a standing invitation
to judicial arbitrariness and policy-driven decisionmaking.” 541 U.S. at 557–58 (Scalia, J.,
dissenting).
114. See Bd. of Trustees of the Univ. of Ala. v. Garrett, 531 U.S. 356, 374 (2001) (holding
Title I of the Americans with Disabilities Act unconstitutional insofar as it subjected states to
private suits in federal court); United States v. Morrison, 529 U.S. 598, 627 (2000) (striking
down the Violence Against Women Act); Kimel v. Fla. Bd. of Regents, 528 U.S. 62, 92 (2000)
(striking down the Age Discrimination in Employment Act’s abrogation of state sovereign
immunity). But the last two statutes scrutinized under the test were found to be congruent and
proportional responses to discrimination and thus legitimate exercises of Congressional power.
See Lane, 541 U.S. 509, 533–34 (2004) (upholding Title II of the Americans with Disabilities
Act); Hibbs, 538 U.S. 721, 740 (2003) (upholding the Family and Medical Leave Act of 1993).
115. For an overview of the effects of City of Boerne and its progeny on equal protection
jurisprudence, see generally Robert C. Post & Reva B. Siegel, Equal Protection By Law: Federal
Antidiscrimination Legislation After Morrison and Kimel, 110 YALE L.J. 441 (2000).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
297
It might be argued that it is the uniquely remedial nature of
Section Five enforcement powers that makes proportionality an
appropriate measure of those powers, and that proportionality should
not be similarly used to constrict the legislative power to prescribe
criminal sanctions. The City of Boerne Court emphasized that
Congress has no power to define substantive rights under the
Fourteenth Amendment, but courts have traditionally recognized an
almost completely unfettered legislative power to define the
substantive criminal law. To make this argument, however, is to
assume one’s conclusion, for what is at stake in the debate over
proportionality is, in part, the scope of the power to define
substantive criminal law. Legal academics sometimes speak of
“substantive criminal law” as including only those laws defining
activity as criminal, “criminal procedure” as those laws governing the
apprehension of suspects and determination of guilt, and “the law of
criminal sentencing” as a third category.116 But a penalty is a necessary
(though perhaps not sufficient) condition of any criminal law, and
thus sentences are as much the “substance” of criminal law as the
categories of prohibited behavior.117 As elaborated in Part III,
proportionality principles have implications for both substantive
criminal law and constitutional criminal procedure.
In any event, the Court has enthusiastically embraced
proportionality as a limitation on exercises of state power in one
context much more akin to criminal sentences than Fourteenth
Amendment remedies: punitive damages in civil tort suits.
B. Proportionality and the Pocketbook
There does not seem to be much doubt that when money is at
stake, proportionality matters. For several years, the Supreme Court
116. See, e.g., Kevin C. McMunigal, The Craft of Due Process, 45 ST. LOUIS U. L.J. 477, 480
(2001) (noting “the contrast between the Supreme Court’s jurisprudence in the area of criminal
procedure and roughly contemporaneous developments in other areas of law critical to the
resolution of criminal cases: substantive criminal law, the law of sentencing and the law of
evidence”).
117. The intersection of sentencing and the substantive criminal law is particularly evident
in Robinson v. California, 370 U.S. 660 (1962), in which the Court found a jail sentence of ninety
days to be “cruel and unusual punishment” for the offense of “be[ing] addicted to the use of
narcotics.” Id. at 662, 667. Robinson is considered one of the Supreme Court’s very few
“substantive criminal law” decisions, but importantly, it was the imposition of the sentence that
gave the Court a constitutional basis to strike down the law. See also William J. Stuntz, The
Uneasy Relationship Between Criminal Procedure and Criminal Justice, 107 YALE L.J. 1, 66–67
(1997) (discussing sentencing as part of the substantive criminal law).
032006 02_RISTROPH.DOC
298
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
has reviewed and sometimes invalidated jury awards of punitive
damages in civil tort suits to ensure that the punitive element of the
118
awards is proportional to the offensive conduct being punished. For
much longer, the Court has reiterated the principle that punitive or
exemplary damages should correspond to “the enormity of [the]
offense.”119 And in the criminal context, the Court has found the
Excessive Fines Clause of the Eighth Amendment to include a
120
proportionality requirement for monetary sanctions.
The Court has not suggested that the need for proportionality
review in the context of civil punitive damages is conditioned on a
particular penological theory. To the contrary, the Court has
explicitly recognized proportionality in this context as essential to
protect individual rights—namely, the property interest identified in
the Due Process Clause of the Fourteenth Amendment. Punitive
damage awards are unconstitutional if they are so “excessive as to
121
amount to a deprivation of property without due process of law.”
To protect this property interest, the Court considers three factors
122
first identified in BMW of North America, Inc. v. Gore : “the degree
of reprehensibility of the defendant’s misconduct,” “the disparity
between the actual or potential harm suffered by the plaintiff and the
punitive damages award,” and “the difference between the punitive
damages awarded by the jury and the civil penalties authorized or
imposed in comparable cases.”123
The Court recently applied the Gore test to reverse a $145
million punitive damage award in State Farm Mutual Insurance Co. v.
Campbell,124 and there the distinction between penological purposes
and arguments for proportionality is even more explicit. The Court
noted, almost as an aside, that punitive damages “are aimed at
118. See State Farm Mut. Auto. Ins. Co. v. Campbell, 538 U.S. 408, 429 (2003) (reversing a
judgment for $145 million in punitive damages on the grounds that the award “was neither
reasonable nor proportionate to the wrong committed”); BMW of N. Am., Inc. v. Gore, 517
U.S. 559, 585–86 (1996) (reversing a judgment for $2 million in punitive damages as “grossly
excessive”); TXO Prod. Corp. v. Alliance Res. Corp., 509 U.S. 443, 453, 482 (1993) (affirming a
$10 million punitive damage award against a challenge that the award was “grossly excessive”).
119. Day v. Woodworth, 54 U.S. (13 How.) 363, 371 (1851); see also Gore, 517 U.S. at 575–
76 (citing cases).
120. United States v. Bajakajian, 524 U.S. 321, 334 (1998).
121. Waters-Pierce Oil Co. v. Texas, 212 U.S. 86, 111 (1909).
122. 517 U.S. 559 (1996).
123. State Farm, 538 U.S. 408 at 418 (citing Gore, 517 U.S. at 575).
124. Id. at 408.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
299
125
deterrence and retribution” and may be imposed “to further a
State’s legitimate interests in punishing unlawful conduct and
deterring its repetition.”126 Immediately thereafter, the Court shifted
gears and announced that “it is well established that there are
procedural and substantive constitutional limitations on these
awards.”127 As State Farm presented the argument, the limitations are
based in the Due Process Clause and are independent of theories of
deterrence or retribution. Indeed, when an award is excessive, its
alleged punitive purpose will be deemed irrelevant: “To the extent an
award is grossly excessive, it furthers no legitimate purpose and
constitutes an arbitrary deprivation of property.”128
Money has also generated searching proportionality review in
129
the criminal context. In United States v. Bajakajian, the Court
applied the Excessive Fines Clause of the Eighth Amendment for the
130
first time. “The touchstone of the constitutional inquiry under the
Excessive Fines Clause is the principle of proportionality: The
amount of the forfeiture must bear some relationship to the gravity of
the offense that it is designed to punish.”131 As with the civil
proportionality requirement of Gore and State Farm, there is no
suggestion that proportionality in the context of criminal fines is
linked to a particular theory of punishment. Rather, the Excessive
Fines Clause announces an individual right, traced in Justice
Thomas’s majority opinion all the way back to the Magna Carta and
traditional concerns about abuses of government power.132 To defend
that right, the Bajakajian Court announced a standard of de novo
review for appellate courts considering excessiveness challenges.133
The Court did not offer a general framework for reviewing fines for
excessiveness, but it considered roughly the same kinds of factors
125. Id. at 416.
126. Id. (quoting Gore, 517 U.S. at 568).
127. Id.
128. Id. at 417.
129. 524 U.S. 321 (1998).
130. See id. at 327 (“This Court has had little occasion to interpret, and has never actually
applied, the Excessive Fines Clause.”). The Eighth Amendment provides that “Excessive bail
shall not be required, nor excessive fines imposed, nor cruel or unusual punishments inflicted.”
U.S. CONST. amend. VIII.
131. Bajakajian, 524 U.S. at 334.
132. Id. at 335.
133. Id. at 336 n.10. Just before announcing the standard of de novo review, the majority
opinion reiterated the importance of judicial deference in the context of proportionality review.
Id. at 336.
032006 02_RISTROPH.DOC
300
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
identified in Gore for proportionality review of civil damages and in
134
Solem v. Helm for proportionality review of criminal sentences.
Specifically, the Bajakajian Court considered the gravity of the
offense (determined in part with reference to civil sanctions for
similar conduct), the harm caused by the offense, and the severity of
the punishment.135
The punitive damages and Excessive Fines cases, along with the
Takings Clause exactions cases, indicate strong support for what
might be called property-interest proportionality. In some
circumstances, a harm, injury, or offense will give the state power to
encroach upon private property rights, but the resultant power must
be proportional to the triggering harm, injury, or offense.
Proportionality is required even when the property seizure is punitive
and regardless of the state’s particular punitive purpose. As a few
commentators have now noted, the proportionality review applied to
monetary penalties has not been matched when freedom rather than
money is at stake.136
C. Death, Prison, and Gross Disproportionality
These examples show that proportionality is well-established in
American constitutional doctrine as a device by which to limit
government power when it encroaches on individual property rights,
and as a device to limit congressional power when it encroaches on
the powers of the states. With respect to nonmonetary criminal
sanctions, the role of proportionality is less clear. Most criminal
defendants who have challenged their sentences as disproportionate
have hung their arguments on the Eighth Amendment’s prohibition
134. 463 U.S. 277 (1983). For a more detailed discussion of Solem v. Helm, see infra text
accompanying notes 186–192.
135. Bajakajian, 524 U.S. at 337–40. Since Bajakajian, the Court has not considered another
challenge to a criminal fine under the Excessive Fines Clause.
136. See, e.g., Chemerinsky, supra note 12, at 1062 (“[T]he Court’s decisions provide that
taking away too much money is unconstitutional, but too many years in prison is not.”); Karlan,
supra note 12, at 882 (“So it’s interesting that, having sharply cut back on proportionality review
of criminal sentences, the Court has identified a proportionality principle for criminal
fines . . . .”); Van Cleave, supra note 12, at 200 (“Yet, ironically, the Court has not shown the
same concern about excessiveness and disproportionality [as it has in the monetary realm] when
the punishment is imprisonment, a deprivation of liberty.”); Adam M. Gershowitz, Note, The
Supreme Court’s Backwards Proportionality Jurisprudence: Comparing Judicial Review of
Excessive Criminal Punishments and Excessive Punitive Damages Awards, 86 VA. L. REV. 1249,
1252 (2000) (“It thus appears that the Supreme Court has not only analyzed excessive criminal
punishment claims separately from excessive punitive damages verdicts, but it has also
promulgated different levels of proportionality review for the two areas.”).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
301
of “cruel and unusual punishments.” This strategy is understandable
enough, but in recent years it has had two unfortunate consequences.
First, proportionality analysis in the criminal context often devolves
into inevitably inconclusive historical arguments over the extent to
which a concern about excessiveness was implicit in the original
meaning of the phrase “cruel and unusual punishments.”137 Second,
those who reject historical or originalist approaches have turned to
penological purpose as an anchor to give meaning to the Cruel and
Unusual Punishments Clause.138 These arguments have led the Court
to overlook proportionality as an implication of basic rights to liberty
and life enshrined in the Constitution and in liberal democratic
principles more generally.
The muddle is only about twenty-five years old. From 1910 until
1980, the Court seemed to recognize that the Eighth Amendment
contained not one but two proportionality principles: a libertyinterest proportionality requirement that incursions on liberty (or
life) be proportionate to the offense for which they were imposed,
and an equality-interest proportionality requirement that
punishments be imposed consistently and nonarbitrarily.139 These
proportionality principles were considered to be separate from but
consistent with the Eighth Amendment’s prohibition of certain modes
of punishment, such as torture.
The Court’s first clear statement of a proportionality
requirement in the Eighth Amendment came in Weems v. United
140
States. There, the Court overturned a sentence of fifteen years of
137. See Solem, 463 U.S. at 312 & n.5 (Burger, C.J., dissenting) (noting that “historians and
scholars have disagreed about the Framers’ original intentions” with respect to the Cruel and
Unusual Punishments Clause, and citing sources).
138. See infra notes 172–222 and accompanying text.
139. As Professor Joseph Raz has pointed out, equality is often mistakenly identified as an
independent principle when it is actually the consequence of the general applicability of other
principles. JOSEPH RAZ, THE MORALITY OF FREEDOM 220 (1986) (“Every moral and every
political theory which claims either that it is a complete theory, or even merely that it is
complete regarding some issue, contains a principle of equality in this sense.”). If what I have
called liberty-interest proportionality were universally realized, then the need for equalityinterest proportionality would disappear. Even though the principle of liberty-interest
proportionality contains within it equality-interest proportionality, I find it useful to speak of
both forms of proportionality. The distinction helps explain the different ways the Supreme
Court speaks of proportionality. Moreover, in practice liberty-interest proportionality is hard to
achieve, and courts may use equality-interest proportionality as a proxy for liberty-interest
proportionality. If one is unsure of the demands of liberty-interest proportionality in a particular
case, one can consider what other jurisdictions do in similar cases.
140. 217 U.S. 349 (1910).
032006 02_RISTROPH.DOC
302
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
“cadena” (a combination penalty that involved imprisonment, hard
labor, and permanent loss of civil rights) for making a false entry in
141
an official document. “[I]t is a precept of justice that punishment for
142
crime should be graduated and proportioned to offense.” The
Weems opinion says almost nothing about penal purposes—except to
recognize the possibility that without constitutional limitations, a
state might use cruelty in its “zeal for a purpose, either honest or
sinister.”143 Because even criminal laws with “honest purposes” can be
cruel, the prohibition on cruel and unusual punishment is “essential
144
to . . . the maintenance of individual freedom.”
To support its conclusion that the Eighth Amendment
prohibition of “cruel and unusual punishments” was both an
antitorture provision and a proportionality requirement, the Weems
Court considered the context in which the Bill of Rights was adopted.
Patrick Henry and other supporters of the Eighth Amendment were
motivated by more than a fear of torture; they also sought to protect
individuals against much more mundane abuses of government
power, including the power to punish.
Henry and those who believed as he did would take no chances.
Their predominant political impulse was distrust of power, and they
insisted on constitutional limitations against its abuse. But surely
they intended more than to register a fear of the forms of abuse that
went out of practice with the Stuarts. Surely, their jealousy of power
had a saner justification than that. They were men of action,
practical and sagacious, not beset with vain imagining, and it must
have come to them that there could be exercises of cruelty by laws
145
other than those which inflicted bodily pain or mutilation.
On this reading of history, the Eighth Amendment is consistent with
other constitutional provisions that seek to limit government power
141. Id. at 357, 366–67.
142. Id. at 367.
143. Id. at 373.
144. Id. (quoting Kepner v. United States, 195 U.S. 100, 122 (1904)). The Weems Court
observed that the Philippine bill of rights’ prohibition on cruel and unusual punishment was
“intended to carry to the Philippine Islands those principles of our government which the
President declared to be established as rules of law for the maintenance of individual freedom.”
Id. at 368 (quoting Kepner, 195 U.S. at 124).
145. Id. at 372. The Weems Court went on to note that a proportionality requirement for
penalties is essential when the legislature has near-absolute power to define what activity is
criminal. Id. For a discussion of the interrelationship between defining crime and fixing
sentences, see infra Part III.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
303
146
even in its ordinary modes of operation. But since Weems, the Court
has not again acknowledged as explicitly that the principles of limited
government and the liberty interests of individuals necessitate a
147
proportionality requirement for criminal sentences.
Individual liberty was nonetheless the basis of the Court’s
148
decision in Robinson v. California, which held that legislatures may
not impose any criminal sentence for “status offenses.”149 Now
characterized as a protection of “ordered liberty” through substantive
150
due process, Robinson invalidated a ninety-day jail sentence
imposed for the offense of drug addiction.151 The Court noted that a
ninety-day sentence, “in the abstract,” is clearly not a categorically
prohibited form of punishment, but the Eighth Amendment is not
simply a restriction on modes of punishment.152 Even a short jail
sentence can be cruel and unusual if imposed for a physiological
153
condition rather than an action. Robinson makes clear that the
Eighth Amendment limits not only the methods used to punish, but
also the severity of punishment. When the state seeks to imprison an
individual, reviewing courts must ask what the imprisonment is for—
what conduct might have triggered the power to punish.154
146. Justice (Edward) White dissented in Weems; he argued that the text and history of the
Cruel and Unusual Punishments Clause showed that it prohibited only “illegal” (not
legislatively authorized) punishments or certain “inhuman” bodily punishments. 217 U.S. at
389–96 (White, J., dissenting). Justice Scalia has advanced a similar historical argument. See,
e.g., Harmelin v. Michigan, 501 U.S. 957, 966–85 (1991) (opinion of Scalia, J.). Debates about
the original meaning of the Eighth Amendment are far from settled, however. As noted, the
Weems majority also relied on historical analysis. One leading article on the original meaning of
the Eighth Amendment argues that proportionality was a component of the British ban on
“cruel and unusual punishments,” and if the American Framers left proportionality out of the
Eighth Amendment, they almost surely did so unintentionally. Anthony F. Granucci, “Nor
Cruel and Unusual Punishments Inflicted”: The Original Meaning, 57 CAL. L. REV. 839, 860–65
(1969).
147. The Court reiterated the Weems holding in Solem v. Helm, 463 U.S. 277, 286–92 (1983),
but was less explicit in Solem about the individual liberty interest that underlies the Eighth
Amendment proportionality requirement.
148. 370 U.S. 660 (1962).
149. Id. at 666–67.
150. See, e.g., Jerold H. Israel, Selective Incorporation: Revisited, 71 GEO. L.J. 253, 292 n.313
(1982) (treating Robinson as part of the “ordered liberties” doctrine).
151. Robinson, 370 U.S. at 661, 667.
152. Id. at 667.
153. Id. at 666–67.
154. Later Supreme Court decisions characterize Robinson as a key proportionality
decision. See, e.g., Atkins v. Virginia, 536 U.S. 304, 311 (2002) (quoting Robinson, 370 U.S. at
666–67).
032006 02_RISTROPH.DOC
304
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
A proportionality requirement that fixes a ceiling on penalties
for a specific offense protects the liberty interest of the individual
defendant. In addition, each individual has an interest in being
treated equally by the criminal law. Equality-interest proportionality
155
motivated several of the many opinions filed in Furman v. Georgia,
the 1972 decision invalidating the death penalty as then applied.156
Justice Douglas found the prohibition on cruel and unusual
punishment to forbid “selective and irregular” penalties as well as
“barbaric” ones.157 “[I]t is ‘cruel and unusual’ to apply the death
penalty—or any other penalty—selectively to minorities whose
numbers are few . . . and who are unpopular, but whom society is
willing to see suffer though it would not countenance general
158
application of the same penalty across the board.” Similarly, Justice
Brennan noted that “the very words ‘cruel and unusual punishments’
imply condemnation of the arbitrary infliction of severe
punishments.”159 Justice Stewart agreed: “These death sentences are
cruel and unusual in the same way that being struck by lightning is
cruel and unusual.”160
In Furman, there was majority support for a principle of
equality-interest proportionality and for the conclusion that the death
penalty as then applied was unconstitutional under that principle. But
no majority agreed that the death penalty was per se disproportionate
to any offense (or per se cruel and unusual for other reasons).
Consequently, states passed new capital punishment statutes that
purported to ensure that the penalty would not be arbitrarily applied.
Just four years after Furman, Georgia’s revised statute passed muster
before the Court.161 One of the features of the Georgia statute singled
out for the Court’s approval was its guarantee of appellate review to
ensure that each death sentence was “proportional to other sentences
155. 408 U.S. 238 (1972) (per curiam).
156. Id. at 240–41. Furman was decided 5–4, and each of the nine Justices filed a separate
opinion.
157. Id. at 242, 245 (Douglas, J., concurring).
158. Id. at 245.
159. Id. at 274 (Brennan, J., concurring). Justice Brennan found three principles inherent in
the Cruel and Unusual Punishments Clause: First, “a punishment must not be so severe as to be
degrading to the dignity of human beings.” Id. at 271. Second, “the State must not arbitrarily
inflict a severe punishment.” Id. at 274. Third, “a severe punishment must not be unacceptable
to contemporary society.” Id. at 277.
160. Id. at 309 (Stewart, J., concurring).
161. Gregg v. Georgia, 428 U.S. 153, 207 (1976).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
305
162
Additionally, Gregg v. Georgia
imposed for similar crimes.”
reiterated the general proportionality requirement first announced in
Weems: a “punishment must not be grossly out of proportion to the
163
severity of the crime.”
One year later, proportionality in Georgia’s capital punishment
164
scheme came before the Court a third time in Coker v. Georgia. The
Court granted a writ of certiorari to consider the claim that the death
penalty was an unconstitutionally disproportionate punishment for
the crime of rape. Although the Coker Court engaged in a kind of
equality-interest analysis by looking across the states (and even to
international law165) to see how many other jurisdictions imposed
capital punishment for rape, its decision ultimately rested on the
166
interest of the individual defendant in life and liberty. “We have the
abiding conviction that the death penalty, which ‘is unique in its
severity and irrevocability,’ is an excessive penalty for the rapist who,
as such, does not take human life.”167
The Coker opinion is not as clear as it could be, and it has
prompted considerable confusion about the need to consider
purposes of punishment when interpreting the Eighth Amendment.
In fact, a misreading of Coker’s holding may be one of the most
important factors contributing to the Court’s retreat from
proportionality review in recent years. Coker interpreted Gregg to
hold that a punishment was unconstitutionally excessive “if it (1)
makes no measurable contribution to acceptable goals of punishment
162. Id. at 203. The Court later held—or, at least, strongly suggested—that this sort of
proportionality review of death sentences was not, in fact, a constitutional requirement. Pulley
v. Harris, 465 U.S. 37, 43–44 (1984). For two perspectives on the scope of the Pulley v. Harris
holding, compare Barry Latzer, The Failure of Comparative Proportionality Review of Capital
Cases (with Lessons From New Jersey), 64 ALB. L. REV. 1161, 1170 (2001) (arguing that Pulley
definitely ended the Constitutional requirement of comparative review), with Evan J. Mandery,
In Defense of Specific Proportionality Review, 65 ALB. L. REV. 883, 899 (2002) (arguing that it is
unclear what sort of proportionality review the Court considered and whether it rejected
anything more than a mere procedural requirement).
163. Gregg, 428 U.S. at 173 (citing Trop v. Dulles, 356 U.S. 86, 100 (1958); Weems v. United
States, 217 U.S. 349, 367 (1910)).
164. 433 U.S. 584 (1977).
165. “[O]ut of 60 major nations in the world surveyed in 1965, only 3 retained the death
penalty for rape where death did not ensue.” Id. at 596 n.10.
166. See id. at 593–97 (“These recent events evidencing the attitude of state legislatures and
sentencing juries do not wholly determine this controversy . . . .”).
167. Id. at 598 (quoting Gregg, 428 U.S. at 187 (citation omitted)). Notably, two of the three
defendants in Furman and its companion cases were sentenced to death for rape rather than
murder. Furman v. Georgia, 408 U.S. 238, 252–53 (1972) (Douglas, J., concurring).
032006 02_RISTROPH.DOC
306
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
and hence is nothing more than the purposeless and needless
imposition of pain and suffering; or (2) is grossly out of proportion to
168
the severity of the crime.” This rule clearly identifies two separate
and independent standards for “excessiveness”—failure to serve
penal purposes and gross disproportionality. A showing that a
punishment fails to serve penal purposes is sufficient to render it
unconstitutional, but such a showing is not necessary to show
unconstitutionality on the separate grounds of disproportionality.
Further demonstrating that penal purposes and proportionality were
separate inquiries, the Coker Court acknowledged that legitimate
penal purposes might sometimes dictate disproportionate
punishments. “Because the death sentence is a disproportionate
punishment for rape, it is cruel and unusual punishment within the
meaning of the Eighth Amendment even though it may measurably
serve the legitimate ends of punishment . . . .”169 When proportionality
and penal purpose conflicted, it was proportionality that was to guide
the Court’s Eighth Amendment analysis.
But later decisions seemed to overlook the or in Coker’s rule and
to collapse the two inquiries. As a result, whether a punishment
serves a legitimate penal purpose has become an increasingly central
question—now, it may be the only question—in proportionality
170
review. In Enmund v. Florida, the Court reversed a death sentence
for a defendant who participated in a robbery that resulted in murder,
171
but did not himself kill or intend to kill. To decide whether the
death penalty was disproportionate in such a situation, the Court
looked only to penal purpose: “Unless the death penalty when
applied to those in Enmund’s position measurably contributes to
[retribution or deterrence], it is ‘nothing more than the purposeless
and needless imposition of pain and suffering,’ and hence an
unconstitutional punishment.”172 More recent decisions have followed
this mode of analysis and treated the question, “Is the punishment
168. Coker, 433 U.S. at 592. The Coker Court referred to Gregg but did not give a page
citation. In fact, Gregg’s statement of the rule does not directly mention penal purposes: “First,
the punishment must not involve the unnecessary and wanton infliction of pain. Second, the
punishment must not be grossly out of proportion to the severity of the crime.” Gregg, 428 U.S.
at 173 (citations omitted).
169. Coker, 433 U.S. 592 n.4 (emphasis added).
170. 458 U.S. 782 (1982).
171. Id. at 788.
172. Id. at 798 (quoting Coker, 433 U.S. at 592).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
307
disproportionate?” as equivalent to the question “Does this
173
punishment serve any legitimate penal purpose?”
The equation of disproportionality with failure to serve a penal
purpose may be one of the most important causes of the Court’s
increased reluctance to conduct proportionality review of criminal
sentences. Penal purpose is widely viewed as a matter of legislative
discretion, and there is great distaste for any judicial review that
requires courts to supervise or second-guess legislatures’ penological
174
aims. Courts are slightly less deferential to legislatures in capital
cases, given the widespread assumption that the unique severity of
death requires more searching judicial review. But the penological
purpose approach to proportionality has undoubtedly produced a
retreat from the notion that the Eighth Amendment prohibits
disproportionately severe prison sentences.
In Rummel v. Estelle,175 the Court upheld a life sentence imposed
under a Texas “recidivist statute” for the commission of three felony
offenses, each of which involved obtaining goods or cash by fraud or
false pretenses.176 The total value of the property involved in all three
felonies was about $230.177 Rejecting the defendant’s argument that a
life sentence was disproportionate to these offenses, the Court noted
that the proportionality review applied in Furman, Gregg, and Coker
was limited to the context of capital punishment. “Because a sentence
of death differs in kind from any sentence of imprisonment, no matter
how long, our decisions applying the prohibition of cruel and unusual
punishments to capital cases are of limited assistance in deciding the
constitutionality of the punishment meted out to Rummel.”178 Justice
Rehnquist’s majority opinion implied that proportionality review had
173. See, e.g., Roper v. Simmons, 125 S. Ct. 1183, 1196–98 (2005) (noting that “the
penological justifications for the death penalty apply to [juveniles] with lesser force than to
adults” and concluding that “the death penalty is disproportionate punishment for offenders
under 18”); Atkins v. Virginia, 536 U.S. 304, 321 (2002) (“We are not persuaded that the
execution of mentally retarded criminals will measurably advance the deterrent or the
retributive purpose of the death penalty . . . . [W]e therefore conclude that such punishment is
excessive . . . .”).
174. See, e.g., Rummel v. Estelle, 445 U.S. 263, 275–76 (1980) (noting that with respect to
the appropriate severity of a criminal penalty, “the lines to be drawn are indeed ‘subjective,’ and
therefore properly within the province of legislatures, not courts”).
175. 445 U.S. 263 (1980).
176. Id. at 264–66.
177. The first offense involved fraudulent use of a credit card to obtain $80 worth of goods
and services. The second offense involved a forged check for $28.36. The third offense involved
obtaining $120.75 by false pretenses. Id. at 265–66.
178. Id. at 272.
032006 02_RISTROPH.DOC
308
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
179
little or no place in the context of prison sentences. (The majority
did acknowledge that a proportionality principle might “come into
play in the extreme example . . . [in which] a legislature made
180
overtime parking a felony punishable by life imprisonment.” ) The
Rummel majority found the Texas statute to be a reasonable exercise
of legislative discretion to pursue legitimate penological objectives.181
The statute’s “primary goals are to deter repeat offenders and, at
some point in the life of one who repeatedly commits criminal
offenses . . . to segregate that person from the rest of society.”182 The
point at which permanent segregation is appropriate is “largely within
183
the discretion of the punishing jurisdiction.” Two years later, in a
184
brief opinion in Hutto v. Davis, the Court reaffirmed Rummel and
upheld a forty-year prison sentence for possession of less than nine
ounces of marijuana.185
A strong proportionality principle in the context of prison
186
sentences made its last, brief stand in Solem v. Helm in 1983. After
stating the general principle that “a criminal sentence must be
proportionate to the crime for which the defendant has been
convicted,”187 Solem announced a three-part test for proportionality
that indicated a concern for both the liberty interests and the equality
188
interests of individual defendants. First, a reviewing court should
consider “the gravity of the offense and the harshness of the
179. See id. at 274 (“[O]ne could argue without fear of contradiction by any decision of this
Court that for crimes concededly classified and classifiable as felonies, that is, as punishable by
significant terms of imprisonment in a state penitentiary, the length of the sentence actually
imposed is purely a matter of legislative prerogative.”). Since Rummel, Chief Justice Rehnquist,
Justice Scalia, and Justice Thomas have been the only Justices to state explicitly that the Eighth
Amendment does not require proportionality in criminal sentences. See Ewing v. California, 538
U.S. 11, 32 (2003) (Thomas, J., concurring) (“In my view, the Cruel and Unusual Punishments
Clause of the Eighth Amendment contains no proportionality principle.”); Harmelin v.
Michigan, 501 U.S. 957, 966–85 (1991) (Scalia, J., joined only by Chief Justice Rehnquist)
(arguing that the Framers did not intend the Cruel and Unusual Punishments Clause to require
proportionality).
180. Rummel, 445 U.S. at 274 n.11.
181. Id. at 278, 284–85.
182. Id. at 284.
183. Id. at 285.
184. 454 U.S. 370 (1982).
185. Id. at 371, 374. The Davis Court quoted extensively from the Rummel opinion and
offered few new principles to guide proportionality analysis. Id. at 372–74.
186. 463 U.S. 277 (1983).
187. Id. at 290.
188. Id. at 290–92.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
309
189
penalty.” This prong of the test draws from the principle expressed
in Weems, Robinson, and Coker that the liberty interest of an
individual can be infringed by a punishing state only in proportion to
the individual’s offense. Next, the Solem Court found, “it may be
helpful to compare the sentences imposed on other criminals in the
same jurisdiction.”190 Finally, “courts may find it useful to compare
the sentences imposed for commission of the same crime in other
191
jurisdictions.” These second two prongs of the test consider both
liberty interests and equality interests. If the same jurisdiction
punishes more serious crimes with the same penalty or less serious
penalties, “that is some indication that the punishment at issue may
be excessive.”192
A deeply divided Court stepped away from the Solem test (but
no majority overruled Solem) in Harmelin v. Michigan.193 Five Justices
voted to uphold a mandatory life sentence for a first-time offender’s
194
possession of 672 grams of cocaine. The majority agreed on
relatively few points, among them that “[t]here can be no serious
contention, then, that a sentence which is not otherwise cruel and
unusual becomes so simply because it is ‘mandatory,’”195 and that the
“individualized” proportionality review necessary for capital
sentences was not required for a mandatory sentence of life in
prison.196 Justice Scalia filed a long separate opinion, joined only by
the Chief Justice, to make a historical argument that “Solem was
simply wrong; the Eighth Amendment contains no proportionality
guarantee.”197 But the most damage to the liberal, not-necessarilypenological proportionality principle was probably done by Justice
Kennedy’s concurrence. Justice Kennedy reviewed the Court’s
proportionality jurisprudence and isolated four principles “that give
content to the uses and limits of proportionality review.”198 Three of
these four principles link proportionality to the purposes of
189. Id. at 290–91.
190. Id. at 291.
191. Id. at 291–92.
192. Id. at 291.
193. 501 U.S. 957 (1991).
194. Id. at 994–97 (Scalia, J., joined by Rehnquist, C.J., and O’Connor, Kennedy, and
Souter, JJ.).
195. Id. at 995.
196. Id.
197. Id. at 965, 966–94 (opinion of Scalia, J.).
198. Id. at 998 (Kennedy, J., concurring).
032006 02_RISTROPH.DOC
310
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
punishment—the liberal, political proportionality principle had all but
199
disappeared. From his four principles, Justice Kennedy drew a
“gross disproportionality” standard: “The Eighth Amendment does
not require strict proportionality between crime and sentence.
Rather, it forbids only extreme sentences that are ‘grossly
disproportionate’ to the crime.”200 According to Justice Kennedy,
reviewing courts should first ask whether the sentence is “grossly
disproportionate” to the offense, and if it is, only then apply the latter
two prongs of the Solem test (intra- and interjurisdictional
201
comparisons). Unless the sentence is grossly disproportionate, a
court should defer to legislative choices of how best to pursue
penological goals.202
The last few Supreme Court decisions to grapple with
proportionality in the criminal context have been similarly enmeshed
in considerations of penological purposes. Although the Court has
not necessarily deferred to legislative determinations of whether
capital punishment serves legitimate purposes, it tends to defer to
legislatures when prison sentences are at stake. In Atkins v.
Virginia,203 the Court emphasized that the Eighth Amendment
protects individuals from “excessive sanctions”204 and found the death
penalty unconstitutionally excessive when imposed on any mentally
205
retarded defendant. The Atkins Court appeared to view “excessive”
199. The first principle is that “the fixing of prison terms for specific crimes involves a
substantive penological judgment that, as a general matter, is ‘properly within the province of
legislatures, not courts.’” Harmelin v. Michigan, 501 U.S. 957, 998 (1991) (Kennedy, J.,
concurring) (quoting Rummel v. Estelle, 445 U.S. 263, 275–76 (1980)). Second, “the Eighth
Amendment does not mandate adoption of any one penological theory.” Id. at 999. Third,
different theories of sentencing and different sentencing practices “are the inevitable, often
beneficial, result of the federal structure.” Id. Fourth, “proportionality review by federal courts
should be informed by ‘objective factors to the maximum possible extent.’” Id. at 1000 (quoting
Rummel, 445 U.S. at 274–75).
200. Id. at 1001 (Kennedy, J., concurring) (quoting Solem v. Helm, 463 U.S. 277, 288
(1983)).
201. Id. at 1005 (Kennedy, J., concurring).
202. Id. Justice Kennedy did not specifically limit this revised Solem test to noncapital
sentences, but his invocations of the value of deference to the legislature focus on the noncapital
context. See id. at 998 (Kennedy, J., concurring) (“[T]he fixing of prison terms . . . involves a
substantive penological judgment that, as a general matter, is properly with the province of
legislatures, not courts.” (internal quotations omitted) (citation omitted)); id. at 999 (“[M]arked
divergences . . . in the length of prescribed prison terms are the inevitable, often beneficial,
result of the federal structure.”).
203. 536 U.S. 304 (2002).
204. Id. at 311.
205. Id. at 321.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
311
as synonymous with “disproportionate,” relying on earlier
proportionality decisions to demonstrate the right against excessive
206
Penological purpose served as the litmus test for
sanctions.
proportionality: the Court reasoned that because capital punishment
could serve neither deterrence nor retribution when imposed on a
mentally retarded defendant, the death penalty would always be
207
unconstitutionally excessive if imposed on such a defendant.
208
In Ewing v. California, a case factually similar to Rummel in
many respects, the Court upheld a life sentence imposed under a
“three strikes” law on a defendant for theft of golf clubs worth about
$400 each.209 By Ewing, penological purpose had become the starting
point of proportionality review, but as a reason to decline to do much
review at all. The majority opinion began by observing the many
important purposes served by California’s three strikes statute,210 and
then emphasized the importance of deference to the legislature on
211
questions of penological purpose. The Ewing Court then adopted
the framework described in Justice Kennedy’s Harmelin concurrence,
announcing that a “narrow proportionality principle” applied to
“noncapital sentences.”212 Under this narrow principle, Ewing’s
sentence was not “grossly disproportionate” to his current felony and
213
his “long history of felony recidivism.” Again, the concept of
proportionality as a liberal principle independent of penal theory
seems to have been eclipsed completely by judicial deference to
legislative penology. The Court emphasized the need for “proper
deference to the policy judgments that find expression in the
legislature’s choice of sanctions.”214 This claim is a long way from
Coker’s recognition that a sentence may be disproportionate “even
215
though it may measurably serve the legitimate ends of punishment.”
206. Id. at 311.
207. Id. at 318–20.
208. 538 U.S. 11 (2003).
209. Id. at 19–20.
210. Id. at 14.
211. Id. at 24 (“Though three strikes laws may be relatively new, our tradition of deferring
to state legislatures in making and implementing such important policy decisions is
longstanding.”).
212. Id. at 20 (quoting Harmelin v. Michigan, 501 U.S. 957, 996–97 (Kennedy, J., concurring
in part and concurring in judgment)).
213. Id. at 29.
214. Id.
215. Coker v. Georgia, 433 U.S. 584, 592 n.4 (1977).
032006 02_RISTROPH.DOC
312
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
216
Lockyer v. Andrade, a habeas case decided alongside Ewing,
similarly reiterated that “the gross disproportionality principle”
should govern judicial review of criminal sentences.217
The “narrow proportionality” principle—perhaps better
described as the gross disproportionality rule—is limited to
noncapital sentences. Ewing is not mentioned at all in Roper v.
Simmons, the Supreme Court’s recent rejection of capital punishment
for any defendant who was less than 18 years old at the time of the
offense.218 Instead of Ewing’s rule that only “grossly disproportionate”
sentences will be struck down, Atkins’s seemingly broader rule
against “excessive sanctions” serves as the basis for Justice Kennedy’s
219
opinion for the Simmons majority. After stating that juveniles have
lesser culpability than adult defendants, Justice Kennedy turned again
to “penological justifications for the death penalty.”220 “In general we
leave to legislatures the assessment of the efficacy of various criminal
221
penalty schemes,” but such deference is not warranted here:
“neither retribution nor deterrence provides adequate justification
222
for imposing the death penalty on juvenile offenders.”
Again, the reluctance of the Roper majority to defer to legislative
judgments about how best to serve penological purposes is probably
limited to the death penalty context. In noncapital criminal
proportionality decisions of recent years, the Court shows a distaste
for substituting (or appearing to substitute) judicial assessments of
216. 538 U.S. 63 (2003).
217. Id. at 77. The specific question before the Court in Andrade was whether California
state courts, in upholding Leandro Andrade’s prison term of fifty years to life, had acted
“contrary to . . . clearly established federal law.” Id. at 70–71 (citing 28 U.S.C. § 2254(d)(1)).
Justice O’Connor, writing for the majority, observed that “our precedents in this area have not
been a model of clarity.” Id. at 72. All that is “clearly established,” Justice O’Connor found, is
the “gross disproportionality principle” articulated in Justice Kennedy’s Harmelin concurrence.
Id. at 73. The majority found that Andrade’s sentence did not violate the gross
disproportionality principle. Id. at 77.
218. Roper v. Simmons, 125 S. Ct. 1183 (2005).
219. Id. at 1190 (“As the Court explained in Atkins, the Eighth Amendment guarantees
individuals the right not to be subjected to excessive sanctions.”). Justice Kennedy has authored
key discussions of proportionality in several contexts: he authored the Roper majority opinion,
the majority opinion in City of Boerne v. Flores, announcing the “congruence and
proportionality” requirement for Congressional action under Section Five of the Fourteenth
Amendment, and a concurrence in Harmelin v. Michigan that first announced the “gross
disproportionality” test later adopted by the Court in Ewing v. California.
220. Id. at 1186.
221. Id. at 1196.
222. Id.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
313
223
appropriate penal strategy for legislative ones. Although the notion
that “death is different” is often invoked to explain the different
constitutional rules that govern capital sentencing,224 it is not clear that
death’s difference would matter with respect to proportionality
judgments. The Court has not explained why the institutional
competence concerns raised in the context of arguably
disproportionate prison sentences are not equally important in the
context of arguably disproportionate death sentences.
In fact, the worry about institutional competence is misplaced
regardless of whether the sentence at issue is prison or death.
Constitutional proportionality is a requirement of a liberal
government, a consequence of individual interests in life, liberty,
property, and equality—and the judiciary plays an important role in
assessing and protecting those individual interests. Decisions about
constitutional proportionality need not be (and should not be)
225
Neither William James
decisions about penological theory.
223. For example, advocates have raised proportionality challenges to lengthy prison terms,
and especially “life without parole” or LWOP sentences, imposed on juvenile offenders, but
have had very little success. See, e.g., Harris v. Wright, 93 F.3d 581, 583–85 (9th Cir. 1996)
(applying the “gross disproportionality” test to reject an Eighth Amendment challenge to a life
without parole sentence imposed on a juvenile offender). See generally Wayne A. Logan,
Proportionality and Punishment: Imposing Life Without Parole on Juveniles, 33 WAKE FOREST
L. REV. 681, 684 (1998) (noting that proportionality challenges to LWOP sentences for juveniles
have had only “limited” success in state courts and no success in the federal courts).
224. See, e.g., Rummel v. Estelle, 445 U.S. 263, 272 (1980) (“Because a sentence of death
differs in kind from any sentence of imprisonment, no matter how long, our decisions applying
the prohibition of cruel and unusual punishments to capital cases are of limited assistance in
deciding the constitutionality of the punishment meted out to Rummel.”).
225. At the same time, it is unlikely that the Court could enforce constitutional rights with
no conceptual account of what punishment is. The Constitution speaks of punishment on several
occasions: Article I, § 3, clause 7; Article I, § 8, clause 6; Article I, § 8, clause 10; Article III, § 3,
clause 2; amendment VIII; amendment XIII. Consequently, the Court has often had to decide
whether a legal consequence is a punishment within the meaning of the Constitution. See, e.g.,
Hamdi v. Rumsfeld, 542 U.S. 507, 556 (2004) (Scalia, J., dissenting) (“It is unthinkable that the
Executive could render otherwise criminal grounds for detention noncriminal merely . . . by
asserting that it was incapacitating dangerous offenders rather than punishing wrongdoing.”);
United States v. Bajakajian, 524 U.S. 321, 328 (1998) (assessing whether a statute mandating
currency forfeiture inflicted punishment); Kansas v. Hendricks, 521 U.S. 346, 360–66 (1997)
(assessing whether a sex offender civil confinement statute imposed punishment); Austin v.
United States, 509 U.S. 602, 619 (1993) (considering which statutory forfeitures should be
considered punishment); Trop v. Dulles, 356 U.S. 86, 94–97 (1958) (assessing whether a statute
providing for denationalization imposes punishment). But we can, and should, define
punishment without adopting a particular theory of penological purpose. The very notion that a
“punishment” could be “cruel and unusual” indicates that the Eighth Amendment assumes a
positive definition of punishment rather than a normative one—in other words, punishment is
not defined in terms of the purposes that allegedly legitimate it. I develop this argument more
032006 02_RISTROPH.DOC
314
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
Rummel, nor Jerry Buckley Helm, nor Gary Albert Ewing claimed a
right to be punished retributively, or a right to be punished under a
Benthamite framework. Instead, each claimed that the power
exercised over him was an infringement of his liberty interest that
could not be justified by the scope of his offense. The fuss about
penological theories in the Court’s recent proportionality decisions is
a distraction from the underlying claim in each case: that the
particular sanction exceeds the state’s legitimate power, given the
offense that is the predicate for the power.
III. BEYOND PENOLOGY: SKEPTICISM AND CONDUCT
Because both the goals of punishment and the means to achieve
those goals are considered matters of legislative prerogative, the
“penological purpose” approach to a constitutional proportionality
requirement in criminal sentencing has greatly narrowed the scope of
that requirement. This close association of proportionality with the
purposes of punishment is neither desirable nor necessary.226 In the
first place, there is little consensus on the appropriate or legitimate
purposes of punishment. More importantly, to the extent that there is
broad support for certain professed aims of punishment—such as
retribution and deterrence—these purposes are defined at a level of
generality that is unlikely to limit sanctions in any meaningful way. In
this Part, I first elaborate the practical and philosophical
shortcomings of penological proportionality, then examine how the
Court might operationalize the concept of political proportionality by
focusing on the conduct that constitutes a criminal offense. A focus
fully elsewhere. See Alice Ristroph, Sexual Punishments, 15 COLUM. J. GENDER & L. 139,
168–70 (2006).
226. Scholars arguing for a stronger constitutional proportionality requirement have, like
the Court, tended to focus on the purposes of punishment (although academic commentators
are typically much less deferential to legislative judgments about the goals of punishment and
the best means to achieve those goals). For example, Professor Frase argues for a
proportionality analysis that accommodates various penological purposes, including both
retributive and utilitarian goals. See Frase, supra note 12, at 588–90. Frase identifies two
different utilitarian proportionality principles—ends proportionality and means
proportionality—and argues that both should figure into Eighth Amendment analysis. Id. at
592–97. The argument for a more searching means-ends analysis seems unlikely to persuade the
Court unless it also addresses the Court’s concern about institutional competence. In Hutto v.
Davis, 454 U.S. 370 (1982), the Court explicitly rejected an argument for a “less restrictive
means” analysis in proportionality review of criminal sentences. See id. at 373 n.2 (rejecting the
lower court’s “less restrictive means” analysis and noting that this analysis had been implicitly
rejected in Rummel, because “the lines to be drawn are indeed ‘subjective,’ and therefore
properly within the province of legislatures, not courts” (quoting Rummel, 445 U.S. at 275–76)).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
315
on conduct is much more likely to provide a workable proportionality
principle that the Supreme Court would and could enforce. In fact,
conduct-based proportionality appears to underlie existing case law
on the constitutional restrictions on sentencing procedures. More
generally, the constitutional law of criminal procedure provides a
model for limitations on penal power that do not presume any
normative theory of punishment.
Proportionality is hardly a strictly retributive principle; it is
clearly demanded by almost every major theory of punishment,
227
including an array of utilitarian theories. It is equally clear that the
precise demands of proportionality will vary from one theory to
another. Some retributive theories may insist that death is the only
proportional punishment for murder, whereas utilitarian theories
might prescribe a far less severe sentence. Disagreement about where
to draw the parameters of proportional punishment does not in itself
render constitutional proportionality analysis impossible. In the
narrow context of capital punishment, the Supreme Court has
accommodated this disagreement by adopting an “overlapping
consensus”228 approach—sentences that are disproportionate under
any accepted penological theory are unconstitutional.229
This overlapping consensus approach to proportionality has led
the Court to reject the death penalty for certain classes of offenders.230
But it has not produced meaningful proportionality review of prison
sentences, and it is unlikely to do so in the future. Notwithstanding
the efforts of many commentators, including the American Law
231
Institute, to use retributivism as a “limiting principle,” the claim
227.
228.
See supra Part I.
“Overlapping consensus” is John Rawls’s term; the Court does not use it. JOHN
RAWLS, A THEORY OF JUSTICE 387–88 (1971); RAWLS, supra note 81, at 15.
229. See, e.g., Roper v. Simmons, 125 S. Ct. 1183, 1196 (2005) (“[N]either retribution nor
deterrence provides adequate justification for imposing the death penalty on juvenile
offenders.”); Atkins v. Virginia, 536 U.S. 304, 321 (2002) (“We are not persuaded that the
execution of mentally retarded criminals will measurably advance the deterrent or the
retributive purpose of the death penalty.”).
230. Roper, 125 S. Ct. at 1200 (striking down the death penalty as applied to juvenile
offenders); Atkins, 536 U.S. at 321 (striking down the death penalty as applied to mentally
retarded offenders); Enmund v. Florida, 458 U.S. 782, 801 (1982) (striking down the death
penalty as applied to defendants who did not kill or intend to kill); Coker v. Georgia, 433 U.S.
584, 600 (1977) (striking down the death penalty as applied to defendants who committed rape
but not murder).
231. See, e.g., MPC SENTENCING REPORT, supra note 9, at 36–37 (noting that the early
revisions to the MPC borrow from Norval Morris’s theory of “limiting retributivism”); Frase,
supra note 12, at 590 (“[Limiting retribution], emphasizing limits on excessive measures, is
032006 02_RISTROPH.DOC
316
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
that a prison sentence serves retributive purposes is, for the most
part, nonfalsifiable. Most retributivist theories focus on the offender’s
desert, and desert is a highly subjective moral notion that is ill-suited
to serve as a constitutional standard. It is not surprising that the Court
has upheld very lengthy prison terms when the state asserts that the
offender “deserves” such punishment232—how would this claim be
disproved? Of course, the claim that the death penalty serves no
retributive purposes for juveniles or the mentally disabled is highly
contested and perhaps also nonfalsifiable. Thus it is not surprising
that the Court’s determination that certain classes of offenders will
never deserve the death penalty has been decried as the unwarranted
imposition of the Justices’ subjective moral judgments.233
consistent with both the text of the Eighth Amendment and the role of constitutional
guarantees.”); Morris, supra note 72, at 201 (“My view is different: It is that desert is not a
defining principle, but is rather a limiting principle.”).
232. The two California three strikes cases, Ewing and Andrade, are particularly interesting
on the question of desert. The state of California identified retribution as well as deterrence and
incapacitation as purposes of the long mandatory sentences imposed under the three strikes
statute. See, e.g., Cal. Penal Code §667(b) (West 1999) (identifying the purposes of the three
strikes statute, including to “ensure greater punishment”); Respondent’s Brief on the Merits,
Ewing v. California, at 8, 18, 21 (arguing that the California law is justified due to the “enhanced
blameworthiness” and “aggravated . . . culpability” of the repeat offender). The Ewing majority
opinion based its decision to uphold the statute on the California legislature’s alleged decision
to adopt incapacitation as its central penological aim. See Ewing v. California, 538 U.S. 11, 25–26
(2003) (“When the California Legislature enacted the three strikes law, it made a judgment that
protecting the public safety requires incapacitating criminals who have already been convicted
of at least one serious or violent crime. Nothing in the Eighth Amendment prohibits California
from making that choice.”); see also id. at 14 (“California’s three strikes law reflects a shift in the
State’s sentencing policies toward incapacitating and deterring repeat offenders who threaten
the public safety . . . .”). But the notion that the three strikes law was aimed at incapacitation
instead of retribution was apparently supplied to counsel by one of the Justices at oral argument.
See Transcript of Oral Argument, Ewing v. California, at *44–45 (“QUESTION: I would have
thought that your response . . . would have been that . . . it depends on what you want your
penal goals to be. California has decided that disabling the criminal is the most important
thing. . . . QUESTION: I mean, proportionality—you necessarily have to look upon what the
principal objective of the punishment is.”). There is nothing obviously false about the claim that
repeat offenders, after a certain number of triggering offenses, deserve life imprisonment.
California made this claim repeatedly before the Supreme Court decided the three strikes law
was better justified in terms of incapacitation.
233. Justice Scalia expresses this view with characteristic vehemence in his dissenting
opinion in Roper v. Simmons. See 125 S. Ct. at 1230 (Scalia, J., dissenting) (suggesting that the
majority opinion rests on little more than “a show of hands on the current Justices’ current
personal views about penology”). But Justice Scalia does not appear to argue that it is
impossible for judges to determine whether an individual deserves a penalty. Instead, he is as
certain that capital punishment for juvenile offenders will serve penological purposes as the
majority is certain that it will not. “The Court’s contention that the goals of retribution and
deterrence are not served by executing murderers under 18 is . . . transparently false.” Id. at
1225.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
317
Whether a sentence serves deterrent purposes is, in theory, a
more useful inquiry. It is easier to define and measure deterrence
than it is to define and measure desert. And yet deterrence is also
unlikely to produce a proportionality standard with enough bite to
prohibit very lengthy prison sentences. Empirical researchers have
been investigating the deterrent effects of criminal sanctions for
234
years. Whether because this research is inconclusive, because it is
thought to be untrustworthy, or for some other reason, the studies
have had little impact on criminal justice policy and on constitutional
standards for criminal law.235
Penological theory thus produces the following problem:
Although all or almost all theories of punishment demand some form
of proportionality, these theories cannot themselves provide a
constitutional standard for meaningful proportionality review. Rather
than abandoning proportionality as a constitutional limitation on
punishment, or becoming resigned to the very weak proportionality
principle of Ewing, one should consider whether the theory of
political proportionality developed in Part I might serve as the basis
of a constitutional standard. Is political proportionality a principle
capable of producing doctrine? If one were agnostic about the
purposes of punishment or uncommitted to any particular theory of
punishment—if one were a penological skeptic—could one
nevertheless articulate the parameters of proportionality?
What might be called penological skepticism is a view that
accepts the practice of punishment as an unavoidable component of a
political system, but refuses to embrace any particular theoretical
justification for this type of exercise of state power.236 The penological
234. See generally DERYCK BEYLEVELD, A BIBLIOGRAPHY ON GENERAL DETERRENCE
RESEARCH (1980) (collecting statistical research to assess the deterrent effect of punishment).
235. See, e.g., Sara Sun Beale, What’s Law Got to Do with It? The Political, Social,
Psychological and Other Non-Legal Factors Influencing the Development of (Federal) Criminal
Law, 1 BUFF. CRIM. L. REV. 23, 29 (1997) (noting the failure of empirical research to influence
U.S. criminal justice policy); see also Charles S. Lanier & James R. Acker, Capital Punishment,
the Moratorium Movement, and Empirical Questions, 10 PSYCHOL. PUB. POL’Y & L. 577, 591
(2004) (“Although the empirical research concerning whether the death penalty deters
homicide more effectively than life imprisonment has been exhaustive and consistent, its role in
judicial and public policy decisions has been less definitive.”).
236. Many scholars have examined various skeptical traditions in American legal thought—
positivism, legal realism, the legal process school, conventionalism, and critical legal studies are
all products of skepticism in one form or another. See, e.g., Michael Moore, Moral Reality, 1982
WIS. L. REV. 1061, 1064–67 (1982) (describing the influence of skepticism on legal process
theorists, conventionalists, positivists, and law and economics scholars, among others); Richard
A. Posner, The Jurisprudence of Skepticism, 86 MICH. L. REV. 827, 827 (1988) (“The skeptical
032006 02_RISTROPH.DOC
318
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
skeptic is not necessarily a punishment abolitionist, but neither is this
skeptic a champion of any normative justification for punishment.
Penological skepticism need not insist that the various justifications
237
for punishment are “wrong.” In the spirit of all good skeptics, the
penological skeptic simply doubts. As a consequence, the penological
skeptic distrusts the coercive penal power and seeks to limit it, but
does not attempt to base those limitations on a theoretical
justification of punishment.238
Can a skeptic have anything to say about punishment under the
Constitution? In fact, to the extent that there exists a body of
constitutional criminal law, it is based on penological skepticism. I
refer, of course, to constitutional criminal procedure. A skeptical
attitude seems to underlie both the text of the constitutional
provisions that protect the accused and the jurisprudence that has
239
interpreted those provisions. The Fourth Amendment prohibition
of unreasonable searches and seizures assumes no theory of
punishment; nor does the Fifth Amendment privilege against selfincrimination. These provisions, along with the Due Process
requirement that guilt be proved beyond a reasonable doubt and the
Sixth Amendment rights to a jury trial, to confrontation, and to
assistance of counsel, are certainly restrictions on the state’s power to
vein in American thinking about law runs from Holmes to the legal realists to the critical legal
studies movement . . . .”).
237. How would one show that retributivism is “wrong”? One can argue against it on liberal
grounds—for instance, there is a strong argument that even if punishment is morally justified,
the liberal state has no business using physical force to satisfy (highly contested) moral
demands. See Murphy, supra note 89, at 510 (“A complete theory of punishment must concern
itself not merely with the moral desirability of the goals sought by punishment . . . but also with
the equally important question of whether the pursuit of these goals is part of the legitimate
business of the state . . . .”). But for many retributivists, the demand for retribution boils down to
a first principle—people who do bad things deserve to be punished. This first principle is an
article of faith; it cannot be proved or disproved but only shared or rejected.
238. One could also say that penological skepticism is based on a positive account of
punishment—punishments are simply the penalties that the state imposes in response to
crime—rather than a normative account that builds justifying aims into the definition of
punishment. As noted above, the Eighth Amendment’s contemplation that “punishments” can
be “cruel and unusual” certainly seems to imply a positive understanding of punishment. See
supra note 225.
239. At any rate, the moral theories that scholars often find to underlie the rules of
substantive criminal law are not generally associated with the procedural rights of criminal
defendants. See Markus Dirk Dubber, American Plea Bargains, German Lay Judges, and the
Crisis of Criminal Procedure, 49 STAN. L. REV. 547, 553 (1997) (noting that “so-called
substantive criminal law” and “criminal procedure” are relatively isolated in American law, and
that it is only substantive criminal law that “associate[s] itself with moral philosophy as best it
can”).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
319
punish: they mandate that the state may not impose punishment
except when it has satisfied certain procedural requirements. But one
would be hard-pressed to show that any of these constitutional
limitations on the penal power assume that punishment will serve
240
retributive, deterrent, or other purposes.
With respect to substantive limitations on criminal sentences, a
penological skeptic would clearly reject any proportionality principle
based on deterrence or retributive theories. But a skeptic could
embrace political proportionality, a limitation on penal power that
considers punishment as a practice situated within a larger political
system that serves many nonpenological goals. A skeptic could
embrace a proportionality requirement that was an external
constraint on the practice of punishment and hence unrelated to the
purpose of penal practices. The constitutional demand that
individuals must be proven guilty beyond a reasonable doubt before
they are punished is unconcerned with the reasons for which the state
241
punishes; it is an external constraint. Because skepticism seems to
underlie the law of constitutional criminal procedure, one might find
ways to “operationalize” the political proportionality principle in the
constitutional law of sentencing procedures.
In a controversial line of cases that has received even more
attention than the recent Eighth Amendment proportionality
240. The Supreme Court’s musings on retribution and deterrence just don’t appear in its
decisions on constitutional criminal procedure. When those decisions attribute a purpose to the
criminal justice system, that purpose is typically “effective law enforcement.” See, e.g.,
California v. Acevedo, 500 U.S. 565, 574 (1991) (holding that probable cause to search a vehicle
extends to containers found within the vehicle, and rejecting a distinction between the vehicle
and containers within it as an impediment to “effective law enforcement”); Tennessee v.
Garner, 471 U.S. 1, 19 (1985) (“We would hesitate to declare a police practice of long standing
‘unreasonable’ if doing so would severely hamper effective law enforcement.”); Kolender v.
Lawson, 461 U.S. 352, 365 (1983) (suggesting that the Fourth Amendment strikes a balance
between “the public interest in effective law enforcement and the equally public interest in
safeguarding individual freedom and privacy”). Neither federal courts nor, to my knowledge,
academic commentators have attempted to explore or explain in detail what it means to enforce
a law or what it means for enforcement to be effective. “Effective law enforcement” could
encompass retributive or utilitarian accounts of punishment, but it need not embrace either of
those theories. In fact, one could understand “enforcing the law” to be the act of ensuring that
the law’s terms are observed. Many criminal laws do not even contain explicit proscriptions
(such as “do not steal”); instead, they define offenses and provide for penalties for those who
commit those offenses. Thus, to enforce the law could mean no more than to ensure that those
who commit the specified behavior are given the specified penalty. There is no theory of
punishment here, no account of why the state attaches penalties to this behavior or what it
hopes the penalties to accomplish.
241. See the discussions of a guilt requirement, supra notes 31 and 94.
032006 02_RISTROPH.DOC
320
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
decisions, the Supreme Court has interpreted the Sixth Amendment
to require that certain facts determinative of the severity of
242
punishment be found by a jury rather than by a judge. These cases
stand for the general proposition that the Sixth Amendment is
implicated not only by the question of guilt or innocence, but also by
marginal increases in the severity of the penalty. One could read
these cases to demonstrate an implicit assumption of proportionality:
If the Constitution contained no proportionality principle, it would
not be offended by marginal increases in the sentence imposed on a
defendant who has been proven guilty of some criminal conduct. But
regardless whether one adopts this reading of the recent Sixth
Amendment decisions, those decisions provide clues for the
implementation of a political proportionality principle.
One of the first decisions in this line of cases, Apprendi v. New
243
Jersey, established that any fact (other than the fact of a prior
conviction) “that increases the penalty for a crime beyond the
prescribed statutory maximum must be submitted to a jury, and
proved beyond a reasonable doubt.”244 After an incident in which
Charles Apprendi fired on an African American family in Vineland,
New Jersey, Apprendi pled guilty to second-degree possession of a
firearm, an offense with a statutory penalty range of five to ten
years.245 The trial judge held an evidentiary hearing and found by a
preponderance of the evidence that Apprendi’s offense had been
246
motivated by racial bias. Under a New Jersey “hate crime” law that
permitted the imposition of enhanced sentences for racially motivated
crimes, the trial judge sentenced Apprendi to twelve years on the
firearm possession count.247 The Supreme Court found that this
242. This line of cases began in 1998 with Almendarez-Torres v. United States, 523 U.S. 224
(1998), and has since produced six other Supreme Court decisions: Jones v. United States, 526
U.S. 227 (1999); Apprendi v. New Jersey, 530 U.S. 466 (2000); Harris v. United States, 536 U.S.
545 (2002); Ring v. Arizona, 536 U.S. 584 (2002); Blakely v. Washington, 124 S. Ct. 2531 (2004);
and United States v. Booker, 125 S. Ct. 738 (2005). The Court addressed similar issues in Monge
v. California, 524 U.S. 721 (1998), but decided that case primarily on Double Jeopardy Clause
grounds rather than Sixth Amendment grounds.
243. 530 U.S. 466 (2000).
244. Id. at 490. Although the Apprendi majority referred repeatedly to the right to a jury
trial, it rested its analysis on the Fourteenth Amendment Due Process Clause and not explicitly
on the Sixth Amendment. Id. at 469, 476.
245. Id. at 469–70. Apprendi was charged with, and pled guilty to, other offenses as well, but
only the constitutionality of the sentence on the firearm possession count was at issue before the
Supreme Court. Id. at 470, 474.
246. Id. at 470–71.
247. Id. at 471.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
321
sentence violated Apprendi’s due process right to a jury
determination of guilt beyond a reasonable doubt.
Arguably, there is an assumption of proportionality that
underlies the Court’s Apprendi analysis. The Constitution has long
been understood to require that guilt of a “crime” must be proven to
248
a jury beyond a reasonable doubt. In Apprendi, the Court used the
severity of the penalty (or more specifically, a marginal increase in
the severity of the penalty) to determine what constituted a “crime”
and therefore what had to be proved to a jury beyond a reasonable
doubt. Since racial hatred was the basis of an increased sentence,
racial hatred must be (in effect if not in label) an element of the
offense for which Charles Apprendi was punished. Apprendi thus
assumes a necessary relationship between the definition of a crime
and the quantity of the penalty. The principle of proportionality
likewise insists on a relationship between the scope of an offense and
the severity of the penalty. But in most theories of proportionality,
the crime is defined first, and a proportional sentence is then
prescribed. Apprendi reversed that analysis and used increases in
penalties to determine changes in the substantive definition of an
offense.249
I would not overemphasize the demand for proportionality in the
sentencing decisions. Standing alone, the assumption of
proportionality in Apprendi and subsequent sentencing decisions
does not reveal whether the Constitution has any substantive (as
opposed to procedural) proportionality requirement. Read narrowly,
these cases stand only for the proposition that if facts trigger an
248. See Mullaney v. Wilbur, 421 U.S. 684, 704 (1975) (holding that because Maine law
provided that “heat of passion on sudden provocation” reduced murder to manslaughter, the
prosecution was required to prove beyond a reasonable doubt the absence of such passion); In
re Winship, 397 U.S. 358, 364 (1970) (“[W]e explicitly hold that the Due Process Clause protects
the accused against conviction except upon proof beyond a reasonable doubt of every fact
necessary to constitute the crime with which he is charged.”).
249. This reversal of typical proportionality reasoning is most clear in Justice Thomas’s
Apprendi concurrence. Apprendi, 530 U.S. at 499 (Thomas, J., concurring). After noting that
“[t]his case turns on the seemingly simple question of what constitutes a ‘crime,’” id., Justice
Thomas analyzed historical evidence and concluded that “a ‘crime’ includes every fact that is by
law a basis for imposing or increasing punishment,” id. at 518. The Court has extended the
Apprendi analysis to capital sentencing, holding that facts necessary to the imposition of a death
sentence must be found by a jury, Ring v. Arizona, 536 U.S. 584, 609 (2002), and to state and
federal sentencing guideline statutes, holding that mandatory guidelines are unconstitutional
insofar as they allow judges to find “sentencing factors” that increase the penalty beyond the
maximum that could be imposed in the absence of the sentencing factor, United States v.
Booker, 125 S. Ct. 738, 756–57 (2005); Blakely v. Washington, 124 S. Ct. 2531, 2543 (2004).
032006 02_RISTROPH.DOC
322
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
increase in the maximum penalty imposed on an individual, those
facts must be evaluated in accordance with the procedural protections
of the Sixth and Fourteenth Amendments. After all, Apprendi says
nothing about whether two years or fifty years is a more appropriate
enhancement for racial bias. It says only that if racial bias results in
any increase beyond the penalty that could be imposed in the absence
of racial bias, then racial bias must be proven to a jury beyond a
reasonable doubt.
Thus one might adopt the view that any penalty prescribed by
the legislature is constitutionally acceptable as long as the jury
determines all facts relevant to that penalty. Just as the Constitution
provides few if any a priori restrictions on what conduct may be
criminalized, it provides few if any a priori restrictions on what
penalty may be attached to criminalized conduct. This may well be
the view of Justices Thomas and Scalia, given that each voted with the
majority in Apprendi and has also rejected an Eighth Amendment
250
proportionality requirement. Justice Scalia’s majority opinion in
Blakely v. Washington flirts with proportionality concerns, but may
ultimately decide that jury participation eliminates any risk of
disproportionality.251 Consider Justice Scalia’s rejection of “legislative
labeling” as an attempt to evade the requirement that juries
determine guilt:
[One alternative] is that the jury need only find whatever facts the
legislature chooses to label elements of the crime, and that those it
labels sentencing factors—no matter how much they may increase
the punishment—may be found by the judge . . . . [A] judge could
sentence a man for committing murder even if the jury convicted
him only of illegally possessing the firearm used to commit it—or of
making an illegal lane change while fleeing the death scene. Not
even Apprendi’s critics would advocate this absurd result. The jury
could not function as circuitbreaker in the State’s machinery of
justice if it were relegated to making a determination that the
defendant at some point did something wrong, a mere preliminary
to a judicial inquisition into the facts of the crime the State actually
252
seeks to punish.
250. Ewing v. California, 538 U.S. 11, 32 (2003) (Thomas, J., concurring); Apprendi, 530 U.S.
at 498 (Scalia, J., concurring); id. at 499 (Thomas, J., concurring); Harmelin v. Michigan, 501
U.S. 957, 966–94 (1991) (opinion of Scalia, J.). As far as I am aware, no Justice has explicitly
contemplated a Sixth Amendment proportionality requirement.
251. 542 U.S. 296, 308–09 (2004).
252. Id. at 306–07 (citation omitted).
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
323
It is difficult to say what is “absurd” about a murder sentence
(presumably, a very severe sentence) for an illegal lane change if not
its disproportionality. Thus, one could read this passage to evoke a
concern for substantive proportionality—a worry that there is
something necessarily wrong with the imposition of a murder-sized
sentence for an illegal lane change.
One could argue, however, that Justice Scalia is concerned not
with substantive proportionality but with sentences whose
proportionality has not been properly evaluated through legislative
and judicial processes. On this view, judgments about the severity of
criminal sentences must be made by democratically accountable
legislatures and sanctioned by juries. Apprendi itself suggests such an
argument. As the Apprendi majority acknowledged in a footnote,
states could try to satisfy Apprendi’s rule without actually increasing
the facts that prosecutors must prove to juries: “[A] State could,
hypothetically, undertake to revise its entire criminal code . . .
extending all statutory maximum sentences to, for example, 50 years
and giving judges guided discretion as to a few specially selected
factors within that range . . . .”253 But the Apprendi Court was
relatively unworried about such statutory amendments that would
simply raise all maximum penalties. “Among other reasons, structural
democratic constraints exist to discourage legislatures from enacting
penal statutes that expose every defendant convicted of, for example,
weapons possession, to a maximum sentence exceeding that which is,
in the legislature’s judgment, generally proportional to the crime.”254
In other words, the “substantive content of . . . criminal laws,”
including statutory penalty ranges, is subject to a “political check on
potentially harsh legislative action.”255 This reasoning in Apprendi,
combined with Blakely’s description of the jury as a “circuitbreaker,”
might suggest that the only constitutional proportionality
requirement is that decisions about the appropriate scope of criminal
sentences be made by a democratically elected legislature and
implicitly ratified by a criminal jury through its findings of fact.
There are significant flaws in the argument that the majoritarian
process and jury trials will, independently or in tandem, be sufficient
to ensure proportionality in criminal sentencing. In the first case,
253. 530 U.S. at 490 n.16.
254. Id.
255. Id. at 491 n.16 (quoting Patterson v. New York, 432 U.S. 197, 228–29 n.13 (1977)
(Powell, J., dissenting)).
032006 02_RISTROPH.DOC
324
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
criminal justice policy is often dictated by factors other than simple
256
majority preferences. Furthermore, limits on some extremely severe
penalties may be desirable notwithstanding popular support for them,
given that the criminal law is a classic example of an area in which
majorities are likely to disregard the rights and interests of
individuals. With respect to the claim that juries can serve as
circuitbreakers to ensure proportional sentences, it is hard to see
what could prompt the jury to throw the switch and break the circuit.
It is unlikely to be the fact that harsh sentencing consequences will
follow from certain factual findings, for in most cases juries will not
have this information. In most U.S. jurisdictions, jurors are not
informed of the specific penalties that attach to a charged offense.257
In fact, courts often deny defendants’ explicit requests that the jury be
instructed of mandatory minimum penalties.258 Thus, though
democratic and legal processes may provide some protection for
individual proportionality interests, this process-based protection is
inadequate, and the judiciary has a role to play in enforcing a
substantive
(and
political,
nonpenological)
proportionality
requirement.
Even if the recent sentencing cases add no further support for
259
such a substantive requirement, they are nonetheless instructive on
the matter of how to implement it. These sentencing decisions
demonstrate a mode of constitutional analysis that measures crimes,
and their constitutional significance, without any reference to
penological purpose. More specifically, these cases elaborate a
256. Professor Franklin Zimring has argued that even “populist” legislation such as
California’s three strikes law does not exactly originate from a spontaneous groundswell of
majority support. See, e.g., FRANKLIN E. ZIMRING, GORDON HAWKINS, & SAM KAMIN,
PUNISHMENT AND DEMOCRACY: THREE STRIKES AND YOU’RE OUT IN CALIFORNIA 160–61
(2001) (arguing that it is wrong to see public opinion as either “solely an input” or “solely an
output” of the political process that determines crime policy, and suggesting that choices by
political leaders sometimes shape public opinion on crime); see also id. at 166 (“The relative
importance of crime as a government issue will also vary over time, and much of that variance
will be produced by developments in noncrime issues” such as war or economic recession.).
257. See, e.g., Erik Lillquist, The Puzzling Return of Jury Sentencing: Misgivings about
Apprendi, 82 N.C. L. REV. 621, 670 (2004) (“In our system, jurors in general are not given
information about punishments, so all they can do is make assumptions about the likely
sentence.”).
258. For a collection of such cases, see Kristen K. Sauer, Note, Informed Conviction:
Instructing the Jury about Mandatory Sentencing Consequences, 95 COLUM. L. REV. 1232, 1245
n.85 (1995) (citing cases).
259. As explained in Part II, supra, there is precedent for the notion of political
proportionality in other doctrinal areas, including the Court’s exactions jurisprudence, its
limitations on punitive damages, and its pre-1980 Eighth Amendment decisions.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
325
constitutional notion of guilt that is based not on retribution (or any
other penological theory) but on actual conduct and mental state.
Conduct-based guilt long preceded Apprendi, but these recent cases
are unusual in their effort to deconstruct guilt from one generalized
category into more nuanced degrees.
The Constitution does not explicitly mandate that only the guilty
be punished, but it is well settled that the Due Process Clause
260
requires proof of guilt as a condition of punishment. “Guilt,” in
constitutional terms, is neither a strictly psychological state (though it
may contain a mens rea requirement) nor an assessment of moral
blameworthiness.261 It is based on conduct and intent: one is guilty if
one has engaged in proscribed conduct with a specified intent, or if
262
one has failed to engage in required conduct with a specified intent.
Constitutional guilt can be assessed only in reference to a specific
criminal offense, and guilt has not traditionally been assessed in
degrees.263 For any given offense, the accused is either guilty or not
guilty. And, of course, the Sixth and Fourteenth Amendments allow a
defendant to demand that this guilt be proven to a jury rather than
found by a judge.
Apprendi, Blakely, and Booker all looked beyond the simple fact
that the defendant was found guilty of something to determine the
precise offense and thus the relative degree of criminal guilt. A
defendant who is found guilty (in constitutional terms) only of
weapon possession cannot be punished with a sentence reserved for
those who possess weapons and are also racially biased. In short, the
260. Mullaney v. Wilbur, 421 U.S. 684, 699–700 (1975); In re Winship, 397 U.S. 358, 361–62
(1970).
261. This distinction is particularly clear in capital punishment proceedings, which are
divided into a “guilt phase” (at which the jury is instructed to decide whether the defendant
engaged in conduct that constitutes a capital offense) and the “penalty phase,” sometimes called
the “culpability phase” (at which the jury is instructed to determine whether the defendant is
sufficiently morally blameworthy to deserve the death penalty).
262. Whether a crime of omission must have an intent requirement was the issue in Lambert
v. California, 355 U.S. 225 (1957). The Supreme Court suggested that intent or at least
constructive knowledge of a legal duty was indeed a requirement—a legislature could not create
strict liability crimes of omission.
[C]onduct alone without regard to the intent of the doer is often sufficient . . . . But
we deal here with conduct that is wholly passive—mere failure to register. It is unlike
the commission of acts, or the failure to act under circumstances that should alert the
doer to the consequences of his deed.
Id. at 228.
263. Offenses are, of course, divided into degrees: first-degree larceny, second-degree
larceny, and so forth. But with respect to any given offense, guilt is all-or-nothing.
032006 02_RISTROPH.DOC
326
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
legislative definition of offenses and sentencing factors—of relevant
types of conduct and mental states—gives the Court a way to make
comparisons between different degrees of guilt.
Legislative definitions of relevant conduct and intent could also
serve as the basis of constitutional proportionality review, especially
in the context of comparative review as required by Solem v. Helm.
Recall that under the three-part test in Solem, a court must consider
(1) the gravity of the offense and the severity of the penalty, (2)
sentences imposed on other offenders in the same jurisdiction, and
(3) sentences imposed on similar offenders in other jurisdictions.264
The focus on conduct and intent in the Apprendi line of cases helps
clarify how reviewing courts would apply the second and third parts
of this test. The more extensive the criminal conduct, the more
expansive the penal power.265
Of course, this type of offense-to-offense comparison will be
easiest (and least controversial) when offenses identify roughly
similar types of conduct as necessary elements. But this will be the
case most of the time; there are few sui generis crimes. So even
though many criminal offenses are arguably incommensurable—who
is to say whether assault is worse than drug distribution?—courts
doing proportionality review could compare assaults to other assaults
and leave to the legislature the control over the relative severity of
sentences for drug offenses in comparison to assault sentences.
The first factor of the Solem test is, of course, inevitably
subjective to some degree. For those who want proportionality review
to exclude all normative judgments, there is not a wholly satisfactory
answer to this problem. But two observations suggest that the
problem is hardly so acute as to require abandonment of judicial
proportionality review. First, the more fixed, concrete aspects of the
Solem test—intra- and interjurisdictional review—can serve as the
primary basis of proportionality judgments. One way to determine
whether a penalty is too severe for a given offense is to ask whether
the state is willing to impose that penalty on everyone who is
convicted of that offense. If the state is unwilling to impose a penalty
264. 463 U.S. 277, 290–92 (1983); see also supra Part II.C.
265. For a fascinating recent example of a similar approach, see United States v. Angelos, 345
F. Supp. 2d 1227 (D. Utah 2004). Considering a proportionality challenge brought under the
Equal Protection Clause rather than the Eighth Amendment, the district court found a
mandatory sentence enhancement for use of a firearm to be irrational (but, perhaps concerned
with being overruled on appeal, the court ultimately imposed the irrational sentence). Id. at
1263.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
327
uniformly, then it is probably safe to conclude that the penalty is
disproportionately severe.
Second, the admittedly subjective requirement that the severity
of the penalty correspond to the severity of the offense is akin to the
requirements of proportionality in other doctrinal contexts. In
reviewing civil punitive damage awards, the courts consider the
“reprehensibility of the defendant’s conduct” in relation to the
266
damage award. Analogously, the test for exactions under the
Takings Clause requires a reviewing court to consider whether the
exaction is proportionate to the impact of the proposed property
development.267 The requirement of proportionality is no more
objective in these contexts or in the context of congressional power
under the Fourteenth Amendment than it is in the context of criminal
sentences.268 In all of these situations, the nature of liberal democracy
and the need to define limits on governmental power require that
judges conduct proportionality review even in the presence of
metaphysical uncertainties.
It should be reemphasized that political proportionality would
focus on the limits of the power to punish—not on the precise severity
of punishment most appropriate to an individual defendant. An
important consequence of this focus, and an important limitation on
proportionality challenges, is that individual defendants would have
to show that their sentences exceed the state’s power over their
conduct. In other words, successful proportionality challenges would
probably be limited to cases in which defendants could persuade a
court that the upper limit of a statutory penalty range was facially
invalid. A political proportionality principle is an effort to limit penal
power based on the conduct that was criminalized—not based on the
particular characteristics of each individual defendant. Or, put
differently, if proportionality is not a matter of individual desert, then
a defendant could not successfully challenge his sentence by arguing,
“Given my particular circumstances, I don’t deserve this penalty.”
*
*
*
266. BMW of N. Am., Inc. v. Gore, 517 U.S. 559, 575 (1996).
267. See supra notes 99–101.
268. See supra Part II.A., which discusses City of Boerne v. Flores, 521 U.S. 507 (1997) and
subsequent applications of its “congruence and proportionality” test.
032006 02_RISTROPH.DOC
328
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
Individual liberty is a fundamental right, both as a matter of
liberal theory and as a matter of American constitutional doctrine.
Disputes about the definition of liberty abound, but it seems clear
that no matter how the term is defined, incarceration is an
infringement of individual liberty. In the context of incarceration,
however, the “fundamental right” status of liberty seems to be
forgotten or neglected. As Sherry Colb has noted, the right to be free
from incarceration is not protected with the same degree of judicial
scrutiny as other fundamental rights.269 Although freedom from bodily
restraint is classified as one of the Constitution’s “fundamental
270
rights,” violations of which ordinarily incur strict scrutiny, prison
sentences have been subjected to heightened scrutiny only when they
also implicate other fundamental rights, such as First Amendment
speech rights271 or Fourteenth Amendment Equal Protection and Due
Process rights.272 Instead, the courts seem to assume that “if the
government may prevent an activity, because there is no fundamental
right to engage in that activity, then the government may elect . . . to
deprive people of their liberty from incarceration as a penalty for
273
engaging in the activity.” Although incarceration is clearly a
deprivation of the right to be free from bodily restraint, the state need
not satisfy the compelling interest or least restrictive means tests to
justify its choice to incarcerate individuals who engage in particular
activities.
Part of the problem, as suggested in Part I, is that criminal law
and punishment are cordoned off from the rest of liberal
constitutional politics, both as a matter of theory and as a matter of
legal doctrine. Except in rare circumstances, such as analyses of civil
269. Sherry F. Colb, Freedom From Incarceration: Why Is This Right Different From All
Other Rights?, 69 N.Y.U. L. REV. 781, 785 (1994) (“An individual’s interest in being free of
physical confinement is a fundamental right . . . . [T]his fundamental right has been treated
differently from all other fundamental rights under the [Supreme] Court’s jurisprudence.”).
270. See id. at 787–88 (citing Meyer v. Nebraska, 262 U.S. 390, 399 (1923), and other cases
that classify freedom from bodily restraint as part of the “fundamental right” to liberty
protected by the Due Process Clause).
271. See, e.g., Texas v. Johnson, 491 U.S. 397, 412 (1989) (applying “the most exacting
scrutiny” to a Texas criminal statute prohibiting destruction of a venerated object, and reversing
the defendant’s sentence of one year in prison and a $2000 fine for burning an American flag
(quoting Boos v. Barry, 485 U.S. 312, 321 (1988))).
272. See, e.g., Loving v. Virginia, 388 U.S. 1, 11–12 (1967) (applying the “most rigid scrutiny”
to reverse criminal sentences under a Virginia statute that prohibited interracial marriage
(quoting Korematsu v. United States, 323 U.S. 214, 216 (1944))).
273. Colb, supra note 269, at 803.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
329
274
disobedience, those suspected or convicted of breaking the law are
treated as lesser beings without the ordinary rights and liberties of
individuals in a constitutional democracy. The enumeration of
specific protections for criminal defendants in the Bill of Rights,
intended no doubt to ameliorate this problem, may actually aggravate
it: judges, policymakers, and academic commentators tend to focus
obsessively on the rights afforded specifically to criminal defendants
and forget that as persons, these defendants also have the same basic
human rights that are held by persons not under the scrutiny of a
punishing authority. In thinking about the scope of the power to
punish, it is necessary to situate that power in the context of the larger
political system. When a liberal state punishes, it must do so
consistently with the individual rights and interests understood to be
fundamental.
But even as one situates the power to punish in its larger political
context, it is also necessary to disaggregate that power in order to
limit it effectively. Punishing is a complex practice that involves
cooperation among all three branches of government and the
exercises of different powers in each branch. The legislature defines
activity as criminal and prescribes a range of penalties; the judiciary
oversees the adjudication of guilt and the determination of sentence;
the executive prosecutes individual defendants and later imposes the
actual sentence—usually, by incarcerating. Attempts to protect
individual defendants’ rights at one point in the penal process are
likely to be ineffective unless the process is viewed as a whole and the
distinct branches can effectively check one another’s power.
These types of concerns surface in arguments for the
“constitutionalization” of substantive criminal law.275 In fact, almost
thirty years ago, scholars calling for a constitutional substantive
criminal law foresaw the quandary that would result if legislative
274. Civil disobedients get much more respect than “ordinary” criminals. Analyses by legal
scholars and philosophers fully recognize the civil disobedient’s continuing interest in liberty,
even after an intentional violation of the law. For example, some theorists wonder whether one
who breaks the law for ideological purposes has an obligation to submit to punishment. See, e.g.,
A.D. Woozley, Civil Disobedience and Punishment, 86 ETHICS 323, 327 (1976) (“If we cannot
make the moral demand of a civil disobedient that he await punishment, then we cannot refuse
to call a man a civil disobedient on the ground that he does not await his punishment.”).
275. See Hart, supra note 13, at 411 (questioning the wisdom of relying on the “legislature’s
sense of justice” rather than prescribing substantive constitutional limitations “on the kinds of
conduct that can be declared illegal”); Stuntz, supra note 13, at 19 (“The normative argument
that justifies the constitutionalization of criminal procedure justifies substantive constitutional
restraints as well.”).
032006 02_RISTROPH.DOC
330
4/24/2006 12:27 PM
DUKE LAW JOURNAL
[Vol. 55:263
power to define crimes were left unchecked. One problem these
scholars identified was that the procedural protections of the Bill of
Rights could be circumvented through creativity in the definitions of
276
criminal offense. For example, Professor Ronald Allen expressed
the fear that legislatures would engage in “semantic
gamesmanship,”277 declining to label some of the categories of
conduct that they wanted to punish as “elements” in order to avoid
the requirement that prosecutors prove that conduct to a jury beyond
a reasonable doubt. In short, Allen foresaw the problem that Blakely
and Booker finally recognized. But Allen’s analysis also suggests that
Blakely and Booker are incomplete solutions. Not surprisingly,
Allen’s own antidote to “semantic gamesmanship” was a kind of
278
proportionality review. Process-based protections of liberty can be
undermined by a legislature with unfettered power to criminalize
conduct. Accordingly, the procedural protections for proportionality
defended in Apprendi, Blakely, and Booker need to be supplemented
with a substantive proportionality requirement.
CONCLUSION
Judicial reluctance to conduct proportionality review has been
the unfortunate result of a mistaken association between
proportionality and particular penological theories. In fact,
proportionality reflects basic liberal principles that have little to do
with penology. A state incursion into individual liberty must be
justified with specific reference to the conduct that prompts the
incursion. Furthermore, impositions of punishment must adhere to
the general rule that similarly situated individuals are equal before
the law.
Whatever value there may be in criminal laws passed by
democratically elected legislatures, the sanction is inevitably an
interference with the bodily freedom (and sometimes, with the very
276. See, e.g., Ronald Jay Allen, Mullaney v. Wilbur, the Supreme Court, and the Substantive
Criminal Law—An Examination of the Limits of Legitimate Intervention, 55 TEX. L. REV. 269,
291 (1977) (“The only clear directive in Wilbur is that Maine, in formulating its criminal statutes
ran afoul of some unarticulated constitutional requirement.”); John Calvin Jeffries & Paul B.
Stephan III, Defenses, Presumptions, and Burden of Proof in the Criminal Law, 88 YALE L.J.
1325, 1328–33 (1979) (“Winship purported to fix the burden of proof as a matter of
constitutional law. To make the scope of that doctrine depend on legislative allocation of the
burden of proof is to assure the point in issue and thus reduce Winship to a circularity.”).
277. Allen, supra note 276, at 290.
278. Id. at 296–98.
032006 02_RISTROPH.DOC
2005]
4/24/2006 12:27 PM
PROPORTIONALITY
331
biological existence) of an individual. But bodily freedom and
biological existence are fundamental individual rights, and under
ordinary circumstances, liberal governments are expected not to
infringe on these rights and even to act affirmatively to protect them.
Necessary and appropriate as criminal sentences may sometimes be,
they must be meted out in a manner consistent with these basic liberal
principles. It is probably the case that a liberal government cannot
avoid exercising force against the bodies of some of its subjects in
some circumstances. But the power to do so must be constrained if
the punishing state is to remain a liberal state, and the best way to
constrain the penal power may be through a liberal principle of
proportionality.